Securitized Products

Highlights Duration: The global economic recovery is more synchronized than at any time since 2011. This suggests that foreign demand will be less of an impediment to the bond bear market and that Treasury yields will rise once U.S. data start to surprise on the upside. Stay at below-benchmark duration. MBS: Agency MBS option-adjusted spreads have widened significantly and no longer look expensive. With Treasury yields moving higher and mortgage refinancings likely to stay depressed, we advise upgrading MBS from underweight to neutral. Economy & Inflation: The U.S. economic data are starting to outperform beaten-down expectations. Survey data point to further GDP acceleration in the second half of this year and we expect inflation will soon follow growth higher. Feature Chart 12-Factor Treasury Model 2-Factor Treasury Model 2-Factor Treasury Model The relationship between the global breadth of economic growth, the value of the dollar and the outlook for Treasury yields has been a running theme in this publication.1 To summarize, stronger global growth pressures bond yields higher (and vice-versa). But how that growth is distributed across different countries matters as well. For example, if global growth is mostly concentrated in the U.S., then yield spreads will widen between the U.S. and the rest of the world and the dollar will appreciate as money pours in from overseas. Investors then respond to a stronger dollar by downgrading their U.S. growth and rate hike expectations. This caps the upside in long-dated U.S. Treasury yields. Conversely, if global growth is more evenly spread out throughout the world, then the dollar will come under less upward pressure when U.S. growth accelerates and Treasury yields can rise further. We developed a simple two-factor model to show how the trade-off between global growth and the exchange rate impacts the U.S. 10-year Treasury yield (Chart 1). The model uses the Global Manufacturing PMI as its proxy for global growth and a survey of bullish sentiment toward the dollar as its proxy for growth synchronization. So far this year, the Global PMI has moved higher and sentiment toward the dollar has become less bullish. Both developments have bond-bearish implications and our model now pegs fair value for the 10-year Treasury yield at 2.65%, 28 bps above the current 10-year yield. In Sync The Global PMI came in at 53.2 in September, the same as in August, but still a strong reading compared to recent history (Chart 2). But the most stunning detail of the September PMI releases is that 33 out of the 36 countries we track had PMIs above the 50 boom/bust line. As a result, our Global PMI Diffusion Index hit 90% for only the second time since 2011 (Chart 2, panel 1). The elevated reading of our diffusion index leads us to two market related observations. First, stronger growth outside of the U.S. explains why the 10-year Treasury yield is only 8 bps lower than at the start of the year despite U.S. economic data that have severely undershot expectations (Chart 2, bottom panel). Second, it suggests that when U.S. economic data inevitably start to surprise on the upside - a process which is only now beginning (see Economy & Inflation section below) - the dollar will appreciate by less than it would have when our PMI diffusion index was near 50. This removes a huge impediment from the bond bear market. In Chart 3 we see that the recent peak in 7-10 year U.S. bond yields occurred at 2.54% on Dec 16th. On that same date the spread between 7-10 year U.S. bond yields and average 7-10 year yields in the rest of the world was 178 bps, and bullish sentiment toward the dollar was above 80%. With the global recovery now more synchronized than it was last year, we anticipate that by the time U.S. yields take out that prior peak, the yield spread and dollar bullish sentiment will still be lower than they were last December. This means that less foreign capital will be encouraged into the U.S. and yields will rise even further. Chart 2Broad Based Recovery Broad Based Recovery Broad Based Recovery Chart 3Spreads Less Of A Constraint Spreads Less Of A Constraint Spreads Less Of A Constraint Where Is Growth Coming From? Considering the major economic blocs, the biggest change during the past year has been the surging Eurozone PMI (Chart 4). The U.S. PMI is still firmly above the 50 boom/bust line but has actually moderated in 2017. The Japanese PMI is similarly entrenched above 50 and while the Chinese PMI was weak earlier this year, it has rebounded during the past four months. At roughly 20%, China carries the largest weight in the Global PMI. The outlook for the Chinese economy is therefore crucial for the path of bond yields. On that note, while the Chinese PMI has been strong in recent months, a couple of warning signs are beginning to flash (Chart 5). Chart 4Global Manufacturing PMIs Global Manufacturing PMIs Global Manufacturing PMIs Chart 5Chinese Monetary Conditions Chinese Monetary Conditions Chinese Monetary Conditions Commodity prices - which correlate strongly with Chinese PMI - have declined since early September, although they remain above levels seen last year and do not yet pose a major risk. What's more important is that monetary conditions are starting to tighten (Chart 5, panel 2). If tighter monetary conditions persist, then we should expect growth to slow. The mild tightening in monetary conditions that has already occurred will probably lead to some near-term moderation in Chinese growth. But our China Investment Strategy service thinks it's unlikely that monetary conditions will tighten enough to cause a meaningful slowdown.2 Our China strategists note that with GDP growth within the government's target range, inflation exceedingly low and signs that financial excesses have been reigned in, there should not be much appetite for draconian policy tightening. We would also add that the causes of this year's tightening in monetary conditions have been relatively benign. The monetary conditions index shown in Chart 5 has fallen because the trade-weighted RMB is no longer depreciating and because real interest rates have moved a tad higher. Crucially, the RMB has only stabilized, it is not appreciating in trade-weighted terms. Also, the nominal policy rate remains flat at a low level. The increase in real interest rates resulted purely from weaker consumer price inflation. Bottom Line: The global economic recovery is more synchronized than at any time since 2011. This suggests that foreign demand will be less of an impediment to the bond bear market and that Treasury yields will rise once U.S. data start to surprise on the upside. Stay at below-benchmark duration. Buy The News In MBS Last week we upgraded our allocation to Agency MBS from underweight to neutral, noting that spreads had become more attractive during the past few months. In all likelihood this is the result of the market pricing in the wind-down of the Fed's balance sheet.3 With the Fed's plans now well known (and unlikely to change), there is an opportunity to increase MBS exposure from a more attractive starting point. After having sold the rumor, we think it's time to buy the news. The Value Proposition Chart 6OAS Look Attractive OAS Look Attractive OAS Look Attractive To be clear, we are not forecasting stellar excess returns from Agency MBS. But with spreads compressed across the entire U.S. fixed income universe, we would note that the option-adjusted spread (OAS) differential between conventional 30-year Agency MBS and investment grade corporate bonds (in duration-matched terms) has risen back to levels last seen in 2014 (Chart 6). The lagged OAS differential is a decent predictor of relative returns between MBS and corporate credit, and at current levels it suggests that MBS could even outperform corporate bonds at some point during the next 12 months (Chart 6, panel 2). This year's decline in Treasury yields has also biased OAS differentials between MBS and corporate bonds wider. Because of negative convexity, MBS duration is positively correlated with yields (Chart 6, bottom panel). If yields rise from here, as we expect they will, then MBS duration will also extend. This means that MBS OAS will start to appear less and less attractive relative to duration-matched comparables. In other words, MBS are less likely to cheapen relative to other spread product in an environment of rising Treasury yields. The Drivers Of MBS Spreads A simplified formula for excess MBS returns, relative to duration-matched Treasuries, could be written as follows: Excess Return = Starting OAS - Duration*(Change in nominal spread) + 0.5*Convexity*(Change in yield) 2 That is, OAS is the correct measure of MBS carry because it adjusts for expected losses due to prepayments. However, it is the change in the nominal spread (not the OAS) that will determine capital gains and losses during the investment horizon. On that note, we observe that nominal MBS spreads have rarely been tighter during the past 30 years (Chart 7). However, it is also hard for us to see a catalyst for significantly wider nominal spreads during the next 6-12 months. The two factors that correlate most closely with nominal MBS spreads are credit spreads and mortgage refinancings. Chart 7Nominal MBS Spreads Are Driven By Credit Spreads And Refinancings Nominal MBS Spreads Are Driven By Credit Spreads And Refinancings Nominal MBS Spreads Are Driven By Credit Spreads And Refinancings On credit spreads, we have repeatedly outlined why they are unlikely to widen materially in the absence of more significant inflationary pressure.4 As for refis, we are also hard pressed to see much upside for three main reasons: First, changes in mortgage rates are the number one driver of refinancings (Chart 8). Refis only increase when mortgage rates fall, making the proposition of refinancing more attractive. As yields rise during the next 6-12 months, refis will stay low. Second, the distribution of outstanding mortgages across the coupon stack impacts how sensitive refis are to changes in rates. The second panel of Chart 8 shows our measure of "moneyness", aka the dispersion of outstanding mortgages around the current coupon rate.5 Given today's dispersion levels we can calculate that even if the current coupon mortgage rate falls back to its recent low of 2.24%, our measure of moneyness would not get back to its late-2016 peak. For our moneyness indicator to rise back to 2013 levels the current coupon mortgage rate would have to fall all the way to 1.68%. Needless to say, we would characterize that risk as low. Third, the final factor that can impact the pace of mortgage refinancing is the seasoning of outstanding mortgages. Typically, we think of mortgages between 30 and 60 months old as being the most likely to refinance. Given that net mortgage origination was close to zero between 30 and 60 months ago and that mortgage purchase applications were at multi-year lows (Chart 9), most of the outstanding mortgage universe probably falls outside of this zone. Chart 8Refis Will Stay Low Refis Will Stay Low Refis Will Stay Low Chart 9Most Mortgages Are Not Yet Seasoned Most Mortgages Are Not Yet Seasoned Most Mortgages Are Not Yet Seasoned Bottom Line: Agency MBS option-adjusted spreads have widened significantly and no longer look expensive. With Treasury yields moving higher and mortgage refinancings likely to stay depressed, we advise upgrading MBS from underweight to neutral. Economy & Inflation Bring On The Upside Surprises As was alluded to in the opening section of this report, after have disappointed expectations year-to-date, we are just now starting to see U.S. economic data surprise to the upside (see Chart 2). The most recent datapoints that caught our eye were the ISM manufacturing and non-manufacturing PMIs.6 Our inclination is to mostly ignore last Friday's employment report as an outlier due to the recent hurricanes.7 The ISM non-manufacturing survey jumped to 59.8 in September, its highest level since 2005. Taken together with other survey indicators that tend to track GDP growth - the BCA Beige Book Indicator and the BCA Composite New Orders Indicator - the case is quite strong for further GDP acceleration in the third and fourth quarters (Chart 10). Of course the pressing issue for bond markets is whether that growth acceleration translates into higher inflation. On that note, we would suggest that the weak inflation we have seen during the past six months was a reaction to the growth slowdown witnessed in 2015 and the first half of 2016. The stronger ISM manufacturing index, in particular, sends a powerful signal that inflation is poised to put in a bottom (Chart 11). Chart 10Survey Indicators Of U.S. Growth Survey Indicators Of U.S. Growth Survey Indicators Of U.S. Growth Chart 11Inflation Lags Growth Inflation Lags Growth Inflation Lags Growth Bottom Line: The U.S. economic data are starting to outperform beaten-down expectations. Survey data point to further GDP acceleration in the second half of this year and we expect inflation will soon follow growth higher. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "Dollar Watching: Another Update", dated January 31, 2017, available at usbs.bcaresearch.com 2 Please see China Investment Strategy Special Report, "On A Higher Note", dated October 5, 2017, available at cis.bcaresearch.com 3 Please see U.S. Bond Strategy Portfolio Allocation Summary, "Return Of The Trump Trade", dated October 3, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Won't Back Down", dated September 26, 2017, available at usbs.bcaresearch.com 5 For each coupon bucket in the Bloomberg Barclays Conventional 30-year Agency MBS index we calculate the squared deviation between its coupon and the current coupon rate. We then weight those squared differences by the market capitalization of each coupon bucket. 6 These are different than the Markit PMI that is included in our 2-factor Treasury model. 7 Please see BCA Daily Insights, "U.S. Jobs Report: All Noise, No Signal", dated October 6, 2017, available at din.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Chart 1Tax Reform Is A Bear-Steepener Tax Reform Is A Bear-Steepener Tax Reform Is A Bear-Steepener The federal government provided some details about its tax reform proposal last week. Markets reacted immediately, once again starting to price-in the possibility of lower tax rates. A basket of high tax-rate stocks outperformed the S&P 500, although the relative price remains well below the highs reached in the immediate aftermath of the election (Chart 1). Bond markets have also been influenced by the "will they, won't they" tax reform drama. Since tax cuts at this relatively late stage of the economic cycle are widely expected to be inflationary, the slope of the yield curve steepens and long-dated TIPS breakevens widen whenever the passage of a tax bill seems more likely. Our political strategists expect that a tax bill will be passed by the end of Q1 2008, or by early Q2 at the latest.1 All else equal, this will bias TIPS breakevens wider and cause the Treasury curve to steepen. Even in the absence of significant tax legislation we think that TIPS breakevens will widen and the yield curve will steepen as inflation starts to pick up during the next few months. But any fiscal stimulus related to tax reform would certainly expedite the process. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 87 basis points in September, bringing year-to-date excess returns up to 234 bps. The average index option-adjusted spread tightened 9 bps on the month to reach 101 bps. Valuation looks increasingly stretched across much of the corporate bond universe. The 12-month breakeven spread for A-rated corporate bonds has dipped well below its mid-2014 trough and is approaching the minimum value witnessed in the early stages of prior Fed tightening cycles. The same measure for Baa-rated credits fell to 17 bps last month, almost exactly equal to its mid-2014 low. While spreads are somewhat expensive, recent data on profit and debt growth have been positive. We noted in last week's report2 that net leverage declined in the second quarter, breaking a streak of two consecutive increases (Chart 2). In addition, other credit cycle indicators such as the slope of the yield curve and C&I bank lending standards do not yet signal wider spreads. Further declines in leverage will depend on whether profit growth can sustain its recent strength (bottom panel). While some moderation is likely, as of now, our leading profit indicators - particularly the weak dollar and surging manufacturing PMI - suggest that growth will stay firm for the remainder of the year (Table 3). Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Return Of The Trump Trade Return Of The Trump Trade Table 3BCorporate Sector Risk Vs. Reward* Return Of The Trump Trade Return Of The Trump Trade High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 143 basis points in September, bringing year-to-date excess returns up to 526 bps. The index option-adjusted spread tightened 31 bps to end the month at 347 bps, 24 bps above the mid-2014 cycle low. After adjusting for expected default losses, we calculate that the junk index currently offers an excess spread of 213 bps. We would expect a default-adjusted spread at this level to translate into low, but positive, excess returns during the next year. A simple linear regression suggests those excess returns will be on the order of 100 to 200 bps (Chart 3), but with a fairly wide margin for error. The default-adjusted spread incorporates our estimate of default losses for the next 12 months. This estimate currently sits at 1.3%. The estimate is derived from the Moody's baseline forecast of a 2.7% default rate and our own estimate of a 51% recovery rate (bottom panel). The relatively benign default outlook is reinforced by the persistent environment of steady growth and low inflation. Last week's third estimate showed that second quarter GDP growth was 3.1%, well above most estimates of trend. Meanwhile, the St. Louis Fed Price Pressures Measure predicts only a 2% chance that inflation will rise above 2.5% during the next year (panel 3). This combo of steady growth and low inflation will ensure that Fed policy remains sufficiently accommodative to support high-yield bond returns. MBS: Upgrade To Neutral Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 35 basis points in September, bringing year-to-date excess returns up to 26 bps. The conventional 30-year MBS yield rose 10 bps in September, driven by a 19 bps increase in the rate component. This was partially offset by an 8 bps tightening of the option-adjusted spread (OAS), while the compensation for prepayment risk (option cost) narrowed 1 bp. OAS have widened considerably during the past few months. In all likelihood this has been in anticipation of the Fed starting to unwind its MBS portfolio. The result is that MBS no longer look expensive compared to Aaa-rated credit (Chart 4). With more attractive valuations and the Fed's schedule for balance sheet runoff now well known, we think the time is right to edge MBS exposure higher. After having sold the rumor of Fed balance sheet runoff, it is time to buy the news. Arbitrage between MBS and credit should limit how much MBS OAS can widen during the next 6-12 months, even in the face of higher MBS supply. Further, recent spread widening has been helped along by falling mortgage rates and rising refinancings. With Treasury yields and mortgage rates now poised to put in a bottom, refis will also roll over and lend support to the MBS trade (bottom panel). Government-Related: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 26 basis points in September, bringing year-to-date excess returns up to 181 bps. Sovereign bonds outperformed the Treasury benchmark by 93 bps on the month. Foreign Agencies and Local Authority bonds outperformed by 25 bps and 46 bps, respectively. Domestic Agency bonds outperformed by 1 bp and Supranationals outperformed by 3 bps. Year-to-date Sovereign bond outperformance has been spurred by dollar weakness, even though spread differentials are tilted firmly in favor of domestic U.S. credit (Chart 5). But with U.S. economic data just now starting to surprise to the upside, we think the tailwind from a weakening dollar is about to fade. Mexico is the single largest issuer in the Sovereign index, and appreciation in the peso versus the U.S. dollar has been a particularly important driver of Sovereign outperformance this year. However, our Emerging Markets Strategy team now believes that peso appreciation is overdone.3 Mexican growth has been supported by strong exports and a weak currency while domestic demand has been soft. Without a solid foundation from domestic demand, this year's currency appreciation will soon cause inflation to roll over and Mexican interest rates to fall. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 62 basis points in September (before adjusting for the tax advantage). Munis have outperformed the Treasury benchmark by 207 bps, year-to-date. The average Municipal / Treasury (M/T) yield ratio edged up from 84% to 86% in September, but it remains extremely tight relative to its post-crisis trading range (Chart 6). State & local government budgets dodged a bullet when the Graham-Cassidy healthcare reform bill was defeated last month. The bill included a block-grant provision for Medicaid that would have reduced federal government transfer payments, a significant source of state & local government revenue. Last week we also learned more specifics about the federal government's proposed tax reform legislation. While the lower tax rates in the proposal are obviously negative for M/T yield ratios, the impact should be somewhat offset by the elimination of tax deductions, the state & local income tax deduction in particular. Eliminating deductions makes the tax advantage in municipal bonds appear more attractive, irrespective of the tax rate. Most importantly, the municipal bond tax exemption itself appears safe. Of course, it will still be some time before we know the final details of tax reform, which our political strategists expect will be passed by the end of Q1 2018. With the plan still not finalized, M/T yield ratios near post-crisis lows look too complacent. Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve shifted higher in September and steepened out to the 5-year maturity point. The 2/10 Treasury slope steepened 7 bps and the 5/30 slope flattened 9 bps. The market brought a December rate hike back into focus last month following a somewhat stronger CPI inflation report and Fed Chair Janet Yellen's insistence that low inflation will prove transitory. Our 12-month fed funds discounter, which shows the market's expected change in the fed funds rate during the next 12 months, moved up to 40 bps from 19 bps. As discussed in last week's report,4 we tend to agree with Chair Yellen that inflation will soon follow growth indicators higher. The market implication of this thesis is that wider TIPS breakevens will lead to one last bout of curve steepening this cycle. We continue to position for curve steepening via a trade long the 5-year bullet and short a duration-matched 2/10 barbell. This trade has returned 16 bps since inception last December. At present, our fair value model shows that the 5-year bullet is slightly expensive on the curve (Chart 7). Or put differently, that the 2/5/10 butterfly spread is fairly priced for 2 bps of 2/10 curve steepening during the next 6 months.5 We think curve steepening will easily surpass this threshold and maintain our long 5-year, short 2/10 position. TIPS: Overweight Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 39 basis points in September, bringing year-to-date excess returns up to -131 bps. The 10-year TIPS breakeven inflation rate rose 8 bps on the month but, at 1.84%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. CPI beat expectations in August for the first time in several months and, as was discussed in a recent report,6 the bond market was quick to react to even a tentative sign that inflation might have troughed. The market's sensitivity should not be surprising. Leading pipeline indicators of inflation, such as the prices paid and supplier deliveries components of the ISM manufacturing index, suggest that inflation and TIPS breakevens are biased higher (Chart 8). Counter-acting some of the optimism on inflation was the slightly weaker-than-expected August PCE report. While trimmed mean PCE inflation did perk up on a month-over-month basis, the 6-month and 12-month rates of change continue to fall (bottom panel). The 2% inflation target is of utmost importance to the Fed. In our base case scenario there is sufficient inflationary pressure for this target to be achieved with a pace of rate hikes similar to the Fed's median projection. But if that turns out not to be the case, then the Fed will respond with a slower pace of hikes. Either way, long-maturity TIPS breakevens must move higher before the end of the cycle or the Fed will have failed. ABS: Cut To Neutral Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities underperformed the duration-equivalent Treasury index by 3 basis points in September, dragging year-to-date excess returns down to 68 bps. Credit card and auto loan ABS both underperformed the duration-equivalent Treasury benchmark by 2 bps in September, pulling year-to-date excess returns down to 67 bps and 69 bps, respectively. The index option-adjusted spread for Aaa-rated ABS widened 3 bps on the month to reach 39 bps. It remains well below its average pre-crisis level (Chart 9). At 39 bps, the Aaa-rated ABS spread is still 11 bps wider than the average option-adjusted spread for conventional 30-year agency MBS. However, as we observed in last week's report,7 delinquency rates for consumer credit (credit cards, auto loans and student loans) are rising, while mortgage delinquency rates continue to fall. This squares with the message from the Fed's Senior Loan Officer Survey which shows that lending standards are tightening for both credit cards and auto loans (bottom panel). While delinquencies appear to have bottomed, the charge-off rate in credit card ABS collateral pools remains near all-time lows. Meanwhile, net losses in auto loan ABS collateral pools are in a clear uptrend. We continue to prefer Aaa-rated credit card ABS over Aaa-rated auto loan ABS, but are wary that credit card charge-offs will also start to increase in the near future, albeit from very low levels. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-agency Commercial Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 6 basis points in September, dragging year-to-date excess returns down to 110 bps. The index option-adjusted spread for non-agency Aaa-rated CMBS widened 1 bp on the month, but it remains well below its average pre-crisis level. Fundamentally, the commercial real estate space continues to be characterized by tightening lending standards and falling demand (Chart 10) and, outside of the multi-family sector, CMBS delinquencies are trending higher (panel 5). Against this back-drop, spreads are not wide enough to entice us. Agency CMBS: Overweight Agency CMBS underperformed the duration-equivalent Treasury index by 18 basis points in September, dragging year-to-date excess returns down to 62 bps. The average index option-adjusted spread for the Agency CMBS index widened 3 bps on the month to reach 51 bps. This compares favorably to the 39 bps offered by Aaa-rated consumer ABS and the 28 bps offered by conventional 30-year Agency MBS. Especially since multi-family delinquency rates remain very low. Treasury Valuation Chart 11Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.65% (Chart 11). Our 3-factor version of the model (not shown), which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.62%. The Global Manufacturing PMI held flat at 53.2 in September, while bullish sentiment toward the dollar crept higher. This caused our model's fair value to edge lower to 2.65% from 2.67%. The U.S., Eurozone and Japan all saw stronger PMIs in September. While China's PMI dipped slightly (from 51.6 to 51), it remains firmly above the 50 boom/bust line. For further details on our Treasury models please refer to the U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com At the time of publication the 10-year Treasury yield was 2.33%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see Geopolitical Strategy Weekly Report, "Can Equities And Bonds Continue To Rally?", dated September 20, 2017, available at gps.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, "Won't Back Down", dated September 26, 2017, avail-able at usbs.bcaresearch.com 3 Please see Emerging Markets Strategy Weekly Report, "Questions From The Road", dated September 20, 2017, available at ems.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Won't Back Down", dated September 26, 2017, available at usbs.bcaresearch.com 5 For further details on our fair value model please see U.S. Bond Strategy Special Report, "Bullets, Barbells And Butterflies", dated July 25, 2017, available at usbs.bcaresearch.com 6 Please see U.S. Bond Strategy Weekly Report, "The Great Unwind", dated September 19, 2017, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy Weekly Report, "Won't Back Down", dated September 26, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights Beige Book highlights disconnect between inflation words and inflation data. Peak in auto sales is not a harbinger of recession. Capital spending still trending higher. Inflation and inflation surprise will need to move higher before Fed hikes again. Big disconnect between 10-year yield and our fair value model. Feature Disconnect On Inflation Chart 1Beige Book Monitors Support##BR##Fed's Outlook On Economy And Inflation Beige Book Monitors Support Fed's Outlook On Economy And Inflation Beige Book Monitors Support Fed's Outlook On Economy And Inflation The Beige Book released on September 6 supports the Fed's base case outlook for the economy and inflation. It also keeps the Fed on track to begin trimming its balance sheet in September and boost rates by another 25 basis points in December if the CPI and PCE inflation readings turn higher. Our quantitative approach to the qualitative data in the Beige Book points to an acceleration in GDP and inflation, less business unease from a rising U.S. dollar, and ongoing improvement in real estate, both commercial and residential (Chart 1). At 64%, the BCA Beige Book Monitor was still near its cycle highs in September, providing further confirmation that economic growth was sturdy in the first two months of Q3. The Fed noted that "the information included in the report was primarily collected before Hurricane Harvey made landfall on the Gulf Coast." However, there was a mention of the storm's clout based on preliminary assessments of business and banking contacts across several districts. The U.S. dollar should not be much of an issue in the Q3 earnings season, according to the Beige Book. The greenback seems to have faded as a concern for small businesses and bankers, in sharp contrast with 2015 and early 2016 when Beige Book references to a strong dollar surged. The Q3 earnings reporting season will provide corporate managements with another forum to discuss the currency's impact on their operations. The 2% decline in the dollar over the past 12 months suggests that the dollar may even provide a small lift to Q3 results (Chart 1, panel 4). Remarkably, business uncertainty over government policy (fiscal, regulatory and health) has moved lower in 2017. The implication is that the business community is largely ignoring the lack of progress by Washington policymakers on Trump's agenda (Chart 1, panel 5). Echoing the market's disagreement with the Fed on inflation, the big disconnect in the Beige Book showed up in the number of inflation words (Chart 1, panel 3). Expressions of inflation dipped between the July and September reports. That said, a wide disconnect remains between the elevated inflation mentions and the soft readings on CPI and PCE. In the past, increased references to inflation have led measured inflation by a few months, suggesting that the CPI and core PCE may soon turn up. Bottom Line: The Beige Book backs the Fed's assertion that the economy will expand around 2% this year and inflation will mount in the coming months, supporting a gradual removal of policy accommodation. Policy uncertainty in Washington and worries over the dollar seem to be fading. The divide between the quantity of inflation words in the Beige Book and measured inflation remains unresolved. Neither the soft data in the Beige Book nor the hard data on the economy suggest that an economic downturn is nigh. Recession Not Imminent Some investors have concluded that the peak in auto sales, a key component of consumer spending on durable goods, suggests that a recession is imminent (Chart 2). We take a different view. Zeniths in consumer durable goods, followed closely by consumer services, were primary harbingers of economic downturns in the post-WWII period. However, expenditures on autos, light trucks and other durables tend to peak seven quarters before the onset of recession. Consumer spending on nondurable goods and services provide less of a warning, topping out just five and four quarters out, respectively. The implication for investors is that the peak in auto sales suggests that a recession is still several years away (Chart 3, panels 1-4). Chart 2Vehicle Sales May##BR##Have Peaked Vehicle Sales May Have Peaked... Vehicle Sales May Have Peaked... Chart 3Consumer Spending And##BR##Housing Prior To Recessions Consumer Spending And Housing Prior To Recessions Consumer Spending And Housing Prior To Recessions Housing investment provides an even earlier indication that a recession is on the horizon (Chart 3, panel-panel 5). Housing peaked 17 quarters before the start of the 2007 recession and 20 quarters, on average, before the onset of the 2001 and 1991 recession. Since the early 1960s, a crest in housing provided seven quarters of warning before a downturn commenced. While housing's contribution to overall economic growth plunged in Q2, we expect housing to provide fuel for the next few years as pent up demand from the depressed household formation rate since the GFC is worked off. The implication from our upbeat view on housing is that the next recession is still several years away. Bottom Line: We expect the next recession to be triggered by an over aggressive Fed, not by imbalances in one of more segments of the economy. It is premature to say that the economy is headed into recession based on a peak in auto sales. Stay long stocks versus bonds, but we recommend that clients be prudent, paring back any overweight positions and holding some safe-haven assets within diversified portfolios. Business Capital Spending Still Up Elevated readings on capex in the first half of the year should persist into the second half. Corporate managements may be postponing investment decisions until they have more clarity on federal tax policy and the Trump administration's plans for infrastructure investment. In short, corporations continue to struggle with how much and when to spend, rather than whether to invest at all. The key supports for sustained corporate spending stayed in place despite the soft July factory orders report and lackluster C&I loan growth. BCA's model for capex (based on non-residential fixed investment, small business optimism and the speculative-grade default rate) suggests lending is poised to climb on a 12-month basis (Chart 4) despite the softening of C&I loan growth since November 2016. Moreover, the 3.3% month-over-month (m/m) drop in factory orders in July masked an upward revision to orders in June and a substantial 1.0% m/m gain in core orders. Core shipments, which feed directly into GDP, rose 1.2% m/m in July. Almost all of the weakness in orders and shipments in July was linked to a 71% plunge in the volatile aircraft orders segment. BCA's research shows that sustainable capital spending cycles get underway only when businesses see evidence that consumer final demand is on the upswing. Consumer expenditures averaged an above-trend 2.7% in 1H. We anticipate that household spending will continue to improve in the second half of 2017.1 Moreover, recent readings on core durable goods orders and shipments show that the uptrend that began in mid-2016 persists, despite recent monthly wiggles in the data (Chart 5). Chart 4BCA Capex Model Points##BR##To Further Improvement BCA Capex Model Points To Further Improvement BCA Capex Model Points To Further Improvement Chart 5Capital Spending##BR##Remains In An Uptrend Capital Spending Remains In An Uptrend Capital Spending Remains In An Uptrend CEO confidence, still a primary support for capex, recently soared to a 13-year high in Q1, but retreated modestly in Q2. The last reading on this survey was in mid-July, and the dip in sentiment reflects the lack of legislative progress in Washington (Chart 5, top panel). The next CEO survey is set for mid-October. The dip in CEO sentiment in Q2 stands in sharp contrast with the easing of concerns around policy in the Beige Book. Chart 6Surprising Drop In Policy##BR##Uncertainty This Year Surprising Drop In Policy Uncertainty This Year Surprising Drop In Policy Uncertainty This Year Surprisingly, the chaos in Washington during the first eight months of the Trump administration has not led to an increase in economic policy uncertainty (Chart 6). Instead, after rising sharply in the wake of the Brexit vote in mid-2016 and the U.S. presidential election in November, policy uncertainty has ebbed. While uncertainty over economic policy remains elevated relative to the past few years, the concern under Trump is surprisingly subdued. This metric is in line with the Beige Book's assessment of Trump's impact on sentiment. A series of business-friendly legislative wins for the GOP and President Trump would further reduce any qualms. Even so, a failure by Congress to boost the debt ceiling and fund the U.S. government later this month would increase business worries/fears. Late last week, Trump cut a deal with Congressional Democrats to extend the debt ceiling for three months and is in talks to do away with it altogether. Bottom Line: The fundamentals still support solid business spending. However, BCA's positive capex outlook in the U.S. could be blemished if the Republicans fail to deliver on their promises to cut taxes and boost infrastructure spending in the next several months. Inflation Surprise And The Fed Chart 7The Fed Cycle And Inflation Surprise The Fed Cycle And Inflation Surprise The Fed Cycle And Inflation Surprise We expect inflation surprise to move higher, which could spur the Fed to resume its rate hike campaign. A disconnect has opened between economic surprise and inflation surprise.2 In the past 13 years, there have been 15 periods when economic surprise has climbed after a trough. The inflation surprise index temporarily increased in 13 of those episodes. For example, in the aftermath of the oil price peak in the U.S. in mid-2014, both economic surprise and inflation surprise diminished through early 2015 and then began climbing. However, today's inflation surprise index has rolled over while economic surprise has gained. The inflation surprise index escalated during previous tightening regimes when the economy was at full employment and the Fed funds rate was in accommodative territory (Chart 7). The last time those conditions were in place, which was in 2005, the Fed was wrapping up a rate increase campaign that began in mid-2004. Mounting inflation surprise also accompanied most of the Fed's rate increases from mid-1999 through mid-2000 under similar conditions. In late 2015, as the current set of rate hikes commenced, the inflation surprise index was on the upswing, the economy was close to full employment and the Fed funds rate was accommodative. What Does This Mean For The Fed? The above analysis underscores that economic growth is in good shape and it is likely to remain so for the next year at a minimum, barring any nasty shocks. Normally, the positive U.S. (and global) growth backdrop would place upward pressure on bond yields. It has not been the case this time. Investors appear skeptical of the ability of strong economic growth to generate higher inflation. The attitude seems to be "we will believe it when we see it". Some on the FOMC are taking a similar attitude. Lael Brainard, a FOMC governor, presented an interesting speech last week that makes this point. She speculated that inflation has been lower post-Lehman for structural reasons related partly to a drop in long-term inflation expectations. The Fed has been reluctant in the past to even hint that inflation expectations have become unmoored, because that could reinforce the trend, thus making it harder for the Fed to move inflation up to target. Brainard, a voting member of the committee with a dovish bias, argued that unemployment may have to undershoot the full employment level for longer than normal because low inflation expectations will be a persistent headwind. She also implied that the central bank should allow inflation to temporarily overshoot the 2% target. At a minimum, she wants to see evidence of rising inflation and inflation expectations before the Fed delivers the next rate hike. In the past, Brainard's speeches have sometimes heralded shifts in the FOMC's consensus. An example is her December 1, 2015 speech at Stanford.3 It is not clear if this is the case this time, but it does reinforce the view that a strong economy and a falling unemployment rate is not enough to justify another rate hike this year according to the consensus on the FOMC. Bottom Line: Our inflation indicators are pointing mildly up. Nonetheless, timing the upturn in inflation is difficult and the Fed will not hike in December without at least a modest rise in inflation (together with higher inflation expectations). We are short duration because Treasuries are overvalued and market expectations for Fed rate hikes over the next year are overly complacent (see next section). Nonetheless, a rise in yields may not be imminent. Disconnect On Duration The Global Manufacturing PMI reached a more than 6-year high in August, climbing from 52.7 in July to 53.1 last month (Chart 8, panel 3). Meanwhile, bullish sentiment toward the U.S. dollar continues to plunge (Chart 8, bottom panel). Together, these two factors suggest that global growth is accelerating and becoming broader based. BCA's U.S. Bond Strategy service4 views the improving global economic backdrop as an extremely bond-bearish development. A wide global recovery means that when U.S. data turns surprisingly positive, it is less likely that any increase in Treasury yields will be met with an influx of foreign demand and surge in the dollar. Our Treasury model (based on Global PMI and dollar sentiment) currently places fair value for the 10-year Treasury yield at 2.67% (Chart 8, top panel). Moreover, our 3-factor version of the model (which includes the Global Economic Policy Uncertainty Index), puts fair value slightly higher at 2.68% (not shown). Investors should continue to position for a steeper curve by favoring the 5-year bullet versus a duration-matched 2/10 barbell. After adjusting for changes in credit rating and duration over time, the average spread offered by the Bloomberg Barclays corporate bond index is fairly valued relative to similar stages of past business cycles. However, the Aaa-rated portion of the market looks expensive. Further, strong Q2 profit growth likely foreshadows a decline in net leverage. This lengthens the window for corporate bond outperformance. We recommend an overweight in the high-yield market. In the early stages of the previous two Fed tightening cycles (February 1994 to July 1994 and June 2004 to December 2005), the index option-adjusted spread averaged 342 bps and traded in a range between 259 bps and 394 bps. This puts the current junk spread (378 bps) almost in line with the average achieved during other similar monetary conditions (Chart 9). We continue to favor a "buy on the dips"5 approach in the high-yield market. Chart 8Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models Chart 9High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview Regarding high-yield valuation, our estimated default-adjusted spread stands at 245 bps. Historically, this level is consistent with excess returns of just under 3% versus duration-matched Treasuries over the subsequent 12 months. Our estimated default-adjusted spread is based on an expected default rate of 2.6% and recovery rate of 49% (Chart 9, bottom panel). We remain underweight MBSs; While MBS are starting to look more attractive, especially relative to Aaa credit, we think it is still too soon to buy. The Fed will announce the run-off of its balance sheet when it meets later this month. The market has been pricing in this eventuality for most of the year, leading to a significant widening in MBS OAS. More recently, the option cost component of MBS spreads has joined in, widening alongside falling mortgage rates and expectations of rising prepayments. Bottom Line: Rates have tested their post-election lows, but BCA's fair value model suggests a bounce higher, which supports our stocks-over-bonds stance. In terms of U.S. bonds, we favor short duration over long and credit over high quality. MBSs will be hurt more than Treasuries as the Fed begins to shrink its balance sheet. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Ryan Swift, Vice President U.S. Bond Strategy ryans@bcaresearch.com Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com 1 Please see BCA's U.S. Investment Strategy Weekly Report, "The Fed's Third Mandate", July 24, 2017. Available at usis.bcaresearch.com. 2 Please see BCA's U.S. Investment Strategy Weekly Report, "Surprise, Surprise", August 28, 2017. Available at usis.bcaresearch.com. 3 https://www.federalreserve.gov/newsevents/speech/brainard20151201a.htm 4 Please see U.S. Bond Strategy Portfolio Allocation Summary, "The Cyclical Sweet Spot Rolls On," September 5, 2017. Available at usbs.bcaresearch.com. 5 Please see BCA's U.S. Bond Strategy Weekly Report, "Keep Buying Dips," March 28, 2017. Available at usbs.bcaresearch.com.

Highlights Chart 1"Trump Trade" Progress Report "Trump Trade" Progress Report "Trump Trade" Progress Report One of our seven investment themes for 2017, published in a Special Report last December, is that the combination of strong U.S. growth and accommodative Fed policy creates a cyclical sweet spot in which risk assets will outperform. After last week's GDP revisions we now know that real growth averaged 2.1% in the first half of the year, solidly above the Fed's 1.8% estimate of trend. Meanwhile, weak inflation has caused markets to discount an exceptionally shallow path for Fed rate hikes - only 19 bps of rate hikes are priced for the next 12 months. This divergence between growth and inflation is reflected in Treasury yields. The real 10-year yield is 24 bps above its pre-election level, while the compensation for inflation protection is only 5 bps higher (Chart 1). Not surprisingly, the cyclical sweet spot has led corporate bonds to outperform duration-matched Treasuries by 296 bps since the election. The persistence of the cyclical sweet spot leads us to believe that last month's politically-driven spread widening should be seen as an opportunity to increase exposure to corporate bonds. Remain at below-benchmark duration and overweight spread product in U.S. fixed income portfolios. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds underperformed the duration-equivalent Treasury index by 62 basis points in August, dragging year-to-date excess returns down to 146 bps. The average index option-adjusted spread widened 8 bps on the month to reach 110 bps. In last week's report,1 we demonstrated that to properly assess corporate bond valuations it is not sufficient to simply look at the average index spread. We need to adjust for the fact that both the average credit rating and duration of the index change over time. We also need to consider corporate spreads relative to other similar stages of the economic cycle, not relative to long-run averages. In this respect, considering the breakeven spread2 for each credit tier relative to where it traded in the early stages of prior Fed tightening cycles gives us the best sense of the value proposition in corporate bonds. At present, this analysis shows that while Aaa corporate spreads are expensive, the other investment grade credit tiers all appear fairly valued (Chart 2). Corporate profit data for the second quarter was released last week and showed a big jump in our measure of EBITD (panel 4). This makes it extremely likely that net corporate leverage declined in Q2. All else equal, this lengthens the window for corporate bond outperformance Table 3.3 Table 3ACorporate Sector Relative Valuation And Recommended Allocation* The Cyclical Sweet Spot Rolls On The Cyclical Sweet Spot Rolls On Table 3BCorporate Sector Risk Vs. Reward* The Cyclical Sweet Spot Rolls On The Cyclical Sweet Spot Rolls On High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield underperformed the duration-equivalent Treasury index by 67 basis points in August, dragging year-to-date excess returns down to 378 bps. The index option adjusted spread widened 26 bps to end the month at 378 bps, 55 bps above the mid-2014 cycle low. Back in March4 we tested a strategy of buying the High-Yield index relative to Treasuries whenever spreads widened by more than 20 bps in a single month, and then holding the trade for a period of one, two or three months. We found that this "buy the dips" strategy works very well when inflationary pressures are low, but performs poorly when inflation is high and rising. When inflation is low the Fed needs to support the recovery by adopting a more dovish posture whenever financial conditions tighten. With the St. Louis Fed Price Pressures Measure5 at only 6% (Chart 3), we expect a "buy the dips" strategy will continue to work for some time. In terms of valuation, our estimated default-adjusted spread stands at 245 bps. Historically, this level is consistent with excess returns of just under 3% versus duration-matched Treasuries over the subsequent 12 months. Our estimated default-adjusted spread is based on an expected default rate of 2.6%, and an expected recovery rate of 49%. MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 12 basis points in August, dragging year-to-date excess returns down to -9 bps. The conventional 30-year MBS yield fell 13 bps in August, driven by an 18 bps decline in the rate component. This was partially offset by a 4 bps increase in the compensation for prepayment risk (option cost) and a 1 bp widening of the option-adjusted spread (OAS). The Fed is likely to announce the run-off of its balance sheet when it meets later this month. For its part, the market has been pricing-in this eventuality for most of the year, leading to a significant widening in MBS OAS. More recently, the option cost component of MBS spreads has joined in, widening alongside falling mortgage rates and expectations of rising prepayments (Chart 4). In this sense, the Fed's commitment to proceed with balance sheet normalization no matter the outlook for the future pace of rate hikes is doubly negative for MBS spreads. OAS are biased wider as Fed buying exits the market, while low rates encourage faster prepayments and a higher option cost component of spreads. Going forward, the option cost component of spreads will decline as mortgage rates cease their downtrend, but OAS still appear too tight relative to trends in net issuance. Despite robust issuance so far this year and the Fed backing away as a buyer, the conventional 30-year MBS OAS remains well below its pre-crisis mean (panel 2). While MBS are starting to look more attractive, especially relative to Aaa credit (panel 3), we think it is still too soon to buy. Government-Related: Underweight Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 5 basis points in August, bringing year-to-date excess returns up to 154 bps. The Foreign Agency and Local Authority sectors drove the index outperformance in August. Both beat the duration-matched Treasury benchmark by 12 bps. Sovereigns outperformed the benchmark by 3 bps, Supranationals outperformed by 1 bp, and Domestic Agency bonds underperformed by 2 bps. We took a detailed look at the Sovereign index in a recent report,6 both at the aggregate and individual country levels. At the aggregate level, the two main factors we consider when deciding whether to add USD-denominated sovereigns to our portfolio at the expense of domestic U.S. credit are relative valuation and the outlook for the U.S. dollar (Chart 5). At present, relative valuation is skewed heavily in favor of domestic U.S. credit (panel 2). Added to that, given downbeat Fed rate hike expectations, we view further dollar weakness as unlikely on a 6-12 month horizon. Taken together, we continue to favor U.S. credit over USD-denominated Sovereign debt. At the country level, we identified several countries where USD-backed debt appears attractive. We found that Finland, Mexico and Colombia all offer attractive spreads. However, the spread pick-up available in Mexican and Colombian debt is compensation for heightened exchange rate volatility. Finnish debt appears the most attractive on a risk/reward basis. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds underperformed the duration-equivalent Treasury index by 40 basis points in August (before adjusting for the tax advantage). Munis have outperformed the Treasury benchmark by 144 bps, year-to-date. The average Municipal / Treasury (M/T) yield ratio held flat in August, and it remains extremely tight relative to its post-crisis trading range (Chart 6). The M/T yield ratio remains very low despite the fact that state & local government net borrowing continues to rise. Net borrowing increased to $209 billion in Q2, the highest level since the second quarter of last year. Further, the Trump administration appears to be finally tackling the issue of tax reform. While comprehensive tax reform is probably too ambitious, some form of corporate and personal tax cuts seems likely, probably in the first half of next year. Lower tax rates are obviously a negative for municipal bonds, but some of the negative impact could be offset if current tax deductions (such as the deduction of state & local income tax) are removed. All else equal, fewer available tax deductions elsewhere makes the tax exemption of municipal bonds look more attractive. Of course, the municipal bond tax exemption itself could also be threatened, but at least so far this appears less likely. The bottom line is that current M/T yield ratios are far too low given the looming risks of rising state & local government borrowing and looming federal tax cuts. Remain underweight. Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve bull flattened in August. The 2/10 slope flattened 17 bps and the 5/30 slope flattened 2 bps. The market moved to discount an even shallower path for Fed rate hikes in August. At the end of July the market had expected 27 bps of rate hikes during the next 12 months, and that number has now fallen to 19 bps (Chart 7). Consequently, our recommendation to short the July 2018 fed funds futures contract has suffered. The position is now 17 bps in the red, but we continue to believe that the market's expected rate hike path is too benign. From current levels, a position short the July 2018 fed funds futures contract will return 35 bps if there are two hikes between now and next July and 61 bps if there are 3 hikes. We also continue to recommend a position long the 5-year bullet versus a duration-matched 2/10 barbell on the view that the Treasury curve will steepen as inflation and TIPS breakevens move higher. This position has earned 28 bps since initiation last December, but valuation is starting to look less attractive. Our butterfly spread model7 suggests that the 5-year bullet is now slightly expensive compared to the 2/10 barbell (panel 3). Or put differently, that the 2/10 Treasury slope will have to steepen by more than 20 bps during the next 6 months for our trade to earn a positive return. TIPS: Overweight Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS underperformed the duration-equivalent nominal Treasury index by 36 basis points in August, dragging year-to-date excess returns down to -169 bps. The 10-year TIPS breakeven inflation rate fell 6 bps on the month and, at 1.76%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. Despite robust growth, extremely weak realized inflation has caused breakevens to tighten this year. Last week's July PCE release was yet another disappointment. The year-over-year core inflation rate fell from 1.51% to 1.41% and the year-over-year trimmed mean rate fell from 1.68% to 1.64% (Chart 8). However, measures of pipeline inflation pressure such as the supplier deliveries and prices paid components of the ISM Manufacturing survey point towards higher inflation. The supplier deliveries component increased from 55.4 to 57.1 in August (panel 4) while the prices paid component held firm at an elevated 62 (panel 3). Adding it all up, and incorporating the fact that employment growth should stay strong enough to maintain downward pressure on the unemployment rate, we think it is very likely that core inflation will soon reverse course and resume the steady uptrend that began in early 2015. TIPS breakevens will widen alongside. At present, our TIPS Financial model suggests that breakevens are trading in line with other financial market instruments (panel 2). In other words, there is no apparent mis-valuation in breakevens relative to other financial markets, and higher realized inflation is likely required before breakevens move sustainably wider. ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 11 basis points in August, bringing year-to-date excess returns up to 71 bps. Aaa-rated ABS outperformed the benchmark by 10 bps in August, bringing year-to-date excess returns up to 63 bps. Meanwhile, non-Aaa ABS outperformed by 26 bps in August, bringing year-to-date excess returns up to 147 bps. Credit card ABS outperformed the Treasury benchmark by 10 bps in August, bringing year-to-date excess returns up to 69 bps. Auto loan ABS outperformed by 12 bps, bringing year-to-date excess returns up to 71 bps. The index option-adjusted spread for Aaa-rated ABS tightened 4 bps on the month, and remains well below its average pre-crisis level (Chart 9). At 36 bps, the option-adjusted spread for Aaa-rated ABS is now the same as the option-adjusted spread for conventional 30-year Agency MBS. Meanwhile, lending standards are now tightening for both auto loans and credit cards. Further, the New York Fed's Household Debt and Credit Report for the second quarter revealed that "flows of credit card balances into both early and serious delinquencies climbed for the third straight quarter - a trend not seen since 2009."8 While overall credit card charge-offs in ABS collateral pools remain low (panel 4), it is clear that the cyclical winds are shifting against consumer ABS. If the trends of tightening lending standards and rising delinquencies continue, then it will soon be time to reduce consumer ABS exposure, possibly shifting into Agency MBS. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 19 basis points in August, bringing year-to-date excess returns up to 116 bps. The index option-adjusted spread for non-agency Aaa-rated CMBS tightened 2 bps on the month, and is approaching one standard deviation below its average pre-crisis level (Chart 10). The combination of tightening lending standards and weaker demand for commercial real estate (CRE) loans (as evidenced by the Fed's Senior Loan Officer Survey) suggests that credit concerns are starting to mount in the CRE space. Meanwhile, CMBS delinquency rates have leveled-off during the past few months and remain much lower in the multi-family space (panel 5). Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 14 basis points in August, bringing year-to-date excess returns up to 79 bps. The average index option-adjusted spread for the Agency CMBS index held flat at 48 bps on the month. This compares favorably to the 36 bps offered by both Aaa-rated consumer ABS and conventional 30-year Agency MBS. Not only does the Agency CMBS sector continue to offer an attractive spread relative to both consumer ABS and Agency MBS, but its agency guarantee and concentration in the multi-family space (where delinquencies are still low) makes it look particularly attractive. Treasury Valuation Chart 11Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.67% (Chart 11). Our 3-factor version of the model (not shown), which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.68%. The Global Manufacturing PMI rose to 53.1 in August, from 52.7 in July, reaching a 75-month high (panel 3). Meanwhile, bullish sentiment toward the U.S. dollar continues to plunge (bottom panel). Taken together, these two factors suggest that not only is global growth accelerating but that the global economic recovery is increasingly broad based. This is an extremely bond-bearish development. A broad based global recovery means that when U.S. data finally start surprising positively, it is less likely that any increase in Treasury yields will be met with an influx of foreign demand. For further details on our Treasury models please refer to the U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com At the time of publication the 10-year Treasury yield was 2.16%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "Policy Reflections", dated August 29, 2017, available at usbs.bcaresearch.com 2 The 12-month breakeven spread is the basis point widening required over a 12-month period before a corporate bond delivers losses relative to a duration-matched Treasury security. We assume no impact from convexity and calculate the breakeven spread as OAS divided by duration. 3 Please see U.S. Bond Strategy Weekly Report, "Low Inflation And Rising Debt", dated June 13, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Keep Buying Dips", dated March 28, 2017, available at usbs.bcaresearch.com 5 The Price Pressures Measure is a composite indicator which shows the percent chance that PCE inflation will exceed 2.5% during the next 12 months. 6 Please see U.S. Bond Strategy Weekly Report, "The Upside Of A Weaker Dollar", dated August 15, 2017, available at usbs.bcaresearch.com 7 For further details on our models please see U.S. Bond Strategy Special Report, "Bullets, Barbells And Butterflies", dated July 25, 2017, available at usbs.bcaresearch.com 8 https://www.newyorkfed.org/microeconomics/hhdc Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights GFIS Portfolio: The GFIS model bond portfolio has lagged its benchmark index since inception last September and since our previous performance update in April. All of that underperformance can be accounted for this month, however, given the risk-off moves seen in global financial markets. As investors begin to shift their attention away from the current geopolitical blustering over North Korea and back towards the solid global economic upturn, our current tilts should begin to outperform again. Risk Management: We have successfully raised the amount of overall portfolio risk (tracking error) since our last portfolio performance update in April. The tracking error remains below our self-imposed limit of 100bps, however, giving us the ability to make further adjustments to our tilts as opportunities arise. Tactical Overlay: Our Tactical Overlay trades have delivered a positive average return over the past year, led by the current open trades that have produced an average gain of +30bps. Feature In this Special Report, we are presenting a performance update for our Global Fixed Income Strategy (GFIS) model bond portfolio. We did the first such update back in mid-April, and we will continue to publish periodic portfolio reviews going forward. As a reminder to our readers, the GFIS model portfolio is intended to be a tool for us to both communicate and evaluate our fixed income investment recommendations. By putting actual weightings to each of our country and sector calls, against a bond benchmark index with an overall portfolio risk limit, we are aiming to express the convictions of our views in a manner more in line with the actual day-to-day portfolio trade-offs faced by bond managers. The model portfolio is a relatively new addition to the GFIS service, starting only in September 2016, thus the return history is still limited. We have built out several pieces of the GFIS model portfolio framework over the past year, and the process is nearing completion. We now have a custom performance benchmark index that reflects the universe of fixed income sectors that we regularly cover in GFIS (essentially, the Bloomberg Barclays Global Aggregate Index plus riskier fixed income classes like High-Yield corporates). We also have performance measurement metrics and a way to regularly present the portfolio returns, while we have also added a risk management (tracking error) element to help size our relative tilts. The final piece will be to incorporate our corporate bond sector recommendations within the model portfolio, both as a source of potential return and a use of our risk budget (tracking error). We intend to add that final element in the coming weeks. Overall Performance Review: Winners & Losers Chart 1GFIS Model Portfolio Performance GFIS Model Portfolio Performance GFIS Model Portfolio Performance As of August 11th, the GFIS model portfolio has produced a total return of +0.93% (hedged into U.S. dollars) since inception on September 20, 2016 (Chart 1). This has underperformed our custom benchmark index by -14bps. Since our last performance review on April 18th, the model portfolio has lagged the benchmark by -10bps. The portfolio has suffered in the risk-off environment seen so far in August, with a -14bp underperformance seen month-to-date, equal to the entire underperformance since inception. Our core structural positions of maintaining a below-benchmark duration stance, while staying underweight government bonds versus overweight spread product, have all suffered of late (bottom two panels). Our government bond country allocation has been the biggest overall drag on returns (Table 1) since last September (-26bps versus our benchmark). Japan (+5bps) and Spain (+3bps) have been the biggest positive contributors since inception, while Italy, the U.K. and France have a combined underperformance of -31bps. That more than accounts for the entire underperformance of the government bond sleeve of the model portfolio since inception (Chart 2). Since our last portfolio update in April, our government bond allocations have lagged our benchmark index by -29bps. Small gains in Spain and Germany (+2bps each) have been dwarfed by underperformance in the U.S. (-16bps), Italy (-10bps) and France (-5bps). Across almost every country, our below-benchmark duration positioning has translated into a bear-steepening yield curve bias, as we have been recommending substantially reduced exposure to the 10+ year maturity buckets in the major countries (U.S., Germany, France, Italy, and Japan). The bull-flattening of global yield curves between March and June, led by a downturn in inflation expectations, was more than large enough to offset any of the potential benefits from our country allocation. Yield curves did began to bear-steepen in July after the European Central Bank (ECB) sent signals that a tapering of its asset purchase program next year was increasingly likely. That move has quickly reversed this month, however, as financial markets have shifted to a risk-off stance on the back of rising geopolitical tensions on the Korean Peninsula. Table 1A Detailed Breakdown Of The GFIS Model Portfolio A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Chart 2GFIS Model Portfolio Government Bond Performance Attribution By Country A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio The news is better with regards to our global spread product allocations. Those have delivered a total return of +1.41% since last September (beating the benchmark by +12bps) and +0.98% since the last performance review in April (+19bps versus the benchmark). Our allocations to U.S. Investment Grade (IG) and High-Yield (HY) have combined for a +30bps outperformance since September and a +23bps outperformance since April (Chart 3). Euro Area corporate debt has been a modest drag, with the combined allocation to IG and HY debt underperforming by -7bps since September and -3bps since April. Emerging Market corporate debt contributed -2bps of underperformance, while U.K. IG corporates added +1bp of excess return. Chart 3GFIS Model Portfolio Spread Product Performance Attribution A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Among other spread sectors, U.S. Mortgage-Backed Securities (MBS) have generated a -12bps contribution to our excess return, although this entirely came from a period immediately after the inception of our model portfolio (Sept-Nov 2016) where we briefly moved to a tactical overweight stance. We have since maintained a structural underweight posture on U.S. MBS, but this has barely generated any relative performance (-1bp) since our last portfolio review in April. Net-net, the GFIS model portfolio has generally performed in line with where our recommendations are concentrated, both in absolute terms and on a relative basis between sectors. Our below-benchmark stance on overall duration has suffered as the government bond yield curves have exhibited more volatility than trend. At the same time, our structural overweights on global corporate debt, favoring the U.S. over non-U.S. equivalents, have contributed positively to the overall portfolio performance. In Charts 4-7, we show the relative performance of some individual countries and sectors that are part of our GFIS benchmark index. We specifically singled out our major asset allocation calls between sectors made over the past year, with a vertical line drawn at the date when the change was recommended. The data shown in all three charts is the relative performance of each tilt on a duration-adjusted basis and (where applicable) hedged back into U.S. dollars, indexed to 100 at the date of implementation in our model portfolio. Shown this way, we can evaluate the success of the timing of our calls. Our shift to an overweight stance on U.S. corporate debt versus U.S. Treasuries both for IG and HY in the first quarter of this year can be judged a success both in terms of timing and magnitude, with IG outperforming Treasuries by 217bps and HY outperforming by 826bps (Chart 4). Within our HY allocation, we left some performance on the table by concentrating our overweights on the higher-rated credit tiers (bottom panel), but this was a move we felt comfortable with (and still do) as a way of staying a bit up in quality at a time when lower-rated spreads were looking fully valued. In terms of our cross-Atlantic credit allocation, we shifted to an overweight stance on U.S. corporates versus Euro Area equivalents back on January 31st of this year (Chart 5). Since then, U.S. IG has underperformed Euro Area IG by -142bps, but U.S. HY has outperformed by a much larger 581bps. Taken together, these positions have contributed positively to the overall performance of the model portfolio. We continue to like U.S. corporates over Euro Area corporates from a valuation standpoint, thus we are keeping this tilt in the portfolio. Chart 4Our Overweights On##BR##U.S. Corporates Have Done Well Our Overweights On U.S. Corporates Have Done Well Our Overweights On U.S. Corporates Have Done Well Chart 5Our Combined Tilt Towards##BR##U.S. Corporates Has Outperformed Our Combined Tilt Towards U.S. Corporates Has Outperformed Our Combined Tilt Towards U.S. Corporates Has Outperformed With regards to our other major spread sector tilts, our shift to an underweight stance on U.S. MBS versus Treasuries back in November has essentially been a wash (Chart 6). Looking ahead, the combination of unattractive valuations and, more importantly, reduced buying of Agency MBS by the Federal Reserve as it begins to shrink its balance sheet will weigh on MBS performance in the next 6-12 months - we are staying underweight. At the same time, we are maintaining our long-held overweight stance on U.K. IG corporates versus Gilts (bottom panel). The Bank of England will be keeping interest rates unchanged over the next year given mixed readings on U.K. economic growth and the lingering uncertainties over the Brexit negotiations, thus going for the added carry of corporates versus expensive Gilts still makes sense. As for our cross-country government bond allocations, our underweight stance on Italy versus Spain, and our overweight stance on Japan versus Germany, have been volatile while delivering no excess performance (Chart 7). Chart 6Sticking With Our Tilts On##BR##U.S. MBS & U.K. IG Sticking With Our Tilts On U.S. MBS & U.K. IG Sticking With Our Tilts On U.S. MBS & U.K. IG Chart 7Our Cross-Country Government Bond##BR##Tilts Have Been Volatile Our Cross-Country Government Bond Tilts Have Been Volatile Our Cross-Country Government Bond Tilts Have Been Volatile Looking ahead, we continue to expect the global growth backdrop to be supportive of spread product over government debt over the next 6-12 months, particularly with central banks unlikely to shift to a restrictive monetary stance. At the same time, we should soon begin to claw back some of the underperformance of the government bond sleeve of the GFIS model portfolio coming from our below-benchmark duration stance, for several reasons: Our colleagues at BCA's Geopolitical Strategy service do not expect the current standoff between Pyongyang and Washington to devolve into a shooting war, even though the tough talk on both sides will likely continue for some time. As the military tensions begin to subside, this should reverse some of the safe-haven bid for government bonds seen in the past couple of weeks, causing yields to drift higher. The solid global growth backdrop, confirmed by the still-rising trend in leading economic indicators, will continue to force central banks to slowly shift to a less dovish policy stance. U.S. inflation will begin to rebound in the next few months, led by the lagged impact of the U.S. dollar weakness seen in 2017 and continued tightening of the U.S. labor market. This will prompt the Fed to hike rates in December and deliver more hikes in 2018, which is NOT currently priced into U.S. Treasuries. We expect the ECB to soon signal a reduction of the size of its asset purchase program starting in 2018, which will put upward pressure on core Euro Area bond yields, and widen Peripheral European spreads, as the market moves to price in a smaller amount of future bond supply that will be absorbed by the central bank. The combination of modest increases in global inflation, a rebound in investor risk sentiment, and an ECB taper announcement should all place bear-steepening pressures on developed market yield curves (ex-Japan). This will benefit the curve-steepening bias we have in the U.S., Euro Area and U.K., while also supporting our country allocation of a maximum overweight to low-beta Japanese Government Bonds (JGBs). Net-net, we see no reason to alter any of current portfolio tilts at the moment based on any change in our market views. Bottom Line: The GFIS model bond portfolio has lagged its benchmark index since inception last September and since our previous performance update in April. Our overweight credit allocations have performed well but our below-benchmark duration tilts have not. All of that underperformance can be accounted for this month, however, given the risk-off moves seen in global financial markets. As investors begin to shift their attention away from the current geopolitical blustering over North Korea and back towards the solid global economic upturn, our current tilts should begin to outperform again. A Very Brief Comment On Our Risk Management Framework In our prior portfolio update in April, we noted that the initial sizes we placed on the tilts in the GFIS model portfolio proved to be far too small to generate any meaningful outperformance.1 After that, we increased the sizes of our all our existing positions in the portfolio. We later introduced a "risk budget" into our framework that would allow us to measure the tracking error (excess volatility versus the GFIS benchmark index) of our portfolio to ensure that we were taking adequate levels of risk.2 So far, our changes have had the desired effect of raising the tracking error of the portfolio to more realistic levels to try and generate outperformance. The average allocations to our government bond underweights and our spread product overweights have increased since that April portfolio review (Chart 8). This has helped raise the tracking error of the model portfolio to 61bps from 25bps in April (Chart 9). This is still below our risk limit of 100bps of tracking error, giving us room to add positions to the model portfolio if we see opportunities come up. Chart 8We've Increased The Sizes Of##BR##Our Tilts Since April ... A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Chart 9...Which Has Boosted The Tracking##BR##Error Of The Model Portfolio ...Which Has Boosted The Tracking Error Of The Model Portfolio ...Which Has Boosted The Tracking Error Of The Model Portfolio Bottom Line: We have successfully raised the amount of overall portfolio risk (tracking error) since our last portfolio performance update in April. The tracking error remains below our self-imposed limit of 100bps, however, giving us the ability to make further adjustments to our tilts as opportunities arise. Tactical Overlay Bets Have Been Helpful In addition to our GFIS model bond portfolio, we also are running recommended trades in our Tactical Overlay portfolio. These are positions that typically have a shorter-term investment time horizon (0-6 months) than those in the model portfolio. They can also be in less-liquid markets that are not included in the custom bond benchmark index for the model portfolio, like U.S. TIPS or New Zealand government bonds. The Overlay is intended to produce ideas for more tactical traders than portfolio managers, although the trades can also be viewed as a compliment to the model bond portfolio. The performance of our Tactical Overlay can be seen in Table 2 (for our current open trades) and Table 3 (for our past closed trades). We have shown the trade performance going back to the inception date of our model bond portfolio in September 2016, to facilitate apples-for-apples comparisons. We are currently working on developing a trade sizing and risk management framework along the lines of our model portfolio. For now, we can only present average return numbers and not a meaningful cumulative return measure. Table 2The Current Open GFIS Tactical Overlay Trades Are Performing Well A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Table 3The Closed GFIS Tactical Overlay Trades Have Been A Mixed Bag A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Our closed Overlay trades since last September generated only an average total return of a mere +1bp, but this weighed down by a large losing position on shorting Portuguese government bonds versus German Bunds. The average trade return would have been +21bps, on fifteen closed trades, excluding that Portuguese bet. The notable winners were long positions in 10-year French government bonds versus German Bunds (+130bps), a long position on Australian Semi-Government debt versus Federal government debt (+159bps) and a long positon on Korean 5-year government bonds vs. 5-year JGBs on a currency-unhedged basis (+195bps). The other notable loser besides the Portuguese trade was a failed long position on Japanese CPI swaps (-111bps). The current open Overlay trades have performed much better, delivering an average gain of +30bps. 14 of the current 16 open trades have a positive gain, thus the batting average is solid. Notable winners are an overweight on U.S. TIPS versus U.S. Treasuries (+197bps) and our Canada/U.K. 2-year/30-year yield curve box trade (+110bps). The only serious losing trade at the moment is our long position in 5-year New Zealand government bonds versus 5-year German debt (-123bps), although this is the only trade in the table that is currency UN-hedged and is a bet on a stronger New Zealand dollar versus the euro as well as a relative bond spread trade. Net-net, our Tactical Overlay trades have generated a positive average return since last September. In the next few months, we will look to introduce a weighting scheme and risk budget for the Overlay trades to better present these trades as a true complement to our model bond portfolio. Bottom Line: Our Tactical Overlay trades have delivered a positive average return over the past year, led by the current open trades that have produced an average gain of +30bps. Robert Robis, Senior Vice President Global Fixed Income Strategy rrobis@bcaresearch.com 1 Please see BCA Global Fixed Income Strategy Special Report, "An Initial Look At The Performance Of Our Model Bond Portfolio", dated April 18th 2017, available at gfis.bcaresearch.com. 2 Please see BCA Global Fixed Income Strategy Special Report, "Adding A Risk Management Framework To Our Model Bond Portfolio", dated June 20th 2017, available at gfis.bcaresearch.com. The GFIS Recommended Portfolio Vs. The Custom Benchmark Index A Performance Update For Our Model Bond Portfolio A Performance Update For Our Model Bond Portfolio Appendix - Selected Sectors From The GFIS Model Portfolio Appendix 1 Appendix 1 Appendix 2 Appendix 2 Appendix 3 Appendix 3 Appendix 4 Appendix 4 Appendix 5 Appendix 5 Appendix 6 Appendix 6 Appendix 7 Appendix 7 Appendix 8 Appendix 8

Highlights Chart 1Too Close For Comfort Too Close For Comfort Too Close For Comfort The Fed is in the midst of tightening policy, but with inflation still below target it wants to ensure that overall policy settings remain accommodative. In the language of central bankers, the Fed wants to keep the real fed funds rate below its equilibrium level, the level that applies neither upward nor downward pressure to price growth. The equilibrium fed funds rate cannot be calculated with precision, but one popular estimate shows that policy settings are dangerously close to turning restrictive (Chart 1). While an announcement of balance sheet reduction is almost certain to occur next month, with the real fed funds rate so close to neutral, rate hikes are probably on hold until the gap widens. Higher inflation will widen the gap by causing the real fed funds rate to fall, and we are confident that core inflation will rise in the coming months (see page 11 for further details). This will permit the Fed to deliver more than the currently discounted 28 bps of rate increases during the next 12 months. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 60 basis points in July, bringing year-to-date excess returns up to 209 bps. The financial press is littered with stories highlighting extremely unattractive corporate bond valuations, but we think this storyline is exaggerated. In fact, the average spread on the Bloomberg Barclays corporate bond index is somewhat wider than is typically observed in the early stages of a Fed tightening cycle (Chart 2). We calculate that in the early stages of the prior two Fed tightening cycles (February 1994 to July 1994 & June 2004 to December 2005), the index option-adjusted spread averaged 86 bps and traded in a range between 66 bps and 104 bps.1 Viewed in this context, the current spread of 102 bps looks somewhat cheap. That being said, corporate balance sheet health is worse than is typically seen during the early stages of a tightening cycle and this will limit spread compression from current levels. But all in all, excess returns to corporate bonds should be consistent with carry during the next 6-12 months, with higher inflation and tighter Fed policy being pre-conditions for material spread widening. In a recent report2 we showed that bank bonds (both senior and subordinate) still offer a spread advantage compared to other similarly risky sectors (Table 3). Banks also continue to make progress shoring up their balance sheets and the outlook for bank profits is starting to brighten. Table 3ACorporate Sector Relative Valuation And Recommended Allocation* On Hold, But Not For Long On Hold, But Not For Long Table 3BCorporate Sector Risk Vs. Reward* On Hold, But Not For Long On Hold, But Not For Long High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 83 basis points in July, bringing year-to-date excess returns up to 448 bps. The index option-adjusted spread tightened 12 bps to end the month at 352 bps, 8 bps above the 2017 low. We calculate that in the early stages of the prior two Fed tightening cycles (February 1994 to July 1994 & June 2004 to December 2005), the index option-adjusted spread averaged 342 bps and traded in a range between 259 bps and 394 bps. This puts the current junk spread almost in line with the average witnessed during other similar monetary environments. In contrast, the VIX index, which co-moves with junk spreads (Chart 3), is well below levels seen during the early stages of the prior two tightening cycles. The VIX currently sits at 10, and its historical range in similar monetary environments is between 11 and 17, with an average of 13.3 In this way, there would appear to be more room for investment grade corporate bond spreads to tighten than junk spreads, especially on a volatility-adjusted basis. Despite somewhat more stretched valuations than in investment grade, high-yield still offers reasonable compensation relative to expected defaults. At present, our estimated default-adjusted spread is 206 bps, only slightly below its historical average (panel 3). This is based on an expected default rate of 2.8% during the next 12 months and an expected recovery rate of 48% (bottom panel). MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 24 basis points in July, bringing year-to-date excess returns up to 4 bps. The conventional 30-year MBS yield declined 3 bps in July, as a small 1 bp increase in the rate component was offset by a 4 bps tightening of the option-adjusted spread (OAS). The compensation for prepayment risk (option cost) held flat. Index OAS has been in a widening trend since bottoming at 15 bps last September (Chart 4). Since then, MBS have returned 43 bps less than duration-equivalent Treasury securities. The Bloomberg Barclays Aaa-rated Credit index has outperformed Treasuries by 71 bps during that same timeframe. The back-up in OAS reflects, in large part, the market pricing in the upcoming wind-down of the Fed's balance sheet, set to be announced next month. However, we think OAS still have further to widen to catch up with the rising trend in net issuance. According to Flow of Funds data, net MBS issuance totaled $83 billion in the first quarter. If that pace continues for the rest of the year, then 2017 will be the strongest year for MBS issuance since 2009. While higher mortgage rates since the end of 2016 present a drag, at least so far, home sales have not shown much weakness (bottom panel). This is unlike the 2013 taper tantrum when home sales fell sharply following the surge in rates. We are underweight MBS on the expectation that the housing market will remain resilient in the face of higher rates, allowing issuance to continue its uptrend. However, we are closely tracking the spread advantage in MBS compared to Aaa-rated credit which is finally starting to look attractive (panel 3). Government-Related: Underweight Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 42 basis points in July, bringing year-to-date excess returns up to 149 bps. Sovereigns and Local Authorities outperformed the Treasury benchmark by 81 bps and 112 bps, respectively. The low-beta Supranational and Domestic Agency sectors each outperformed by 5 bps. The Foreign Agency sector outperformed the duration-matched Treasury index by 56 bps. USD-denominated sovereign bonds have underperformed the Baa-rated U.S. Corporate index (their closest comparable in terms of risk) during the past three months even though the U.S. dollar has continued its trend lower (Chart 5). But despite this recent underperformance, the Sovereign index still does not offer a spread advantage over the Baa-rated U.S. Corporate index (panel 3). Further, while our Emerging Markets Strategy service still looks favorably upon the Mexican peso relative to other emerging market currencies, it does not expect the peso to continue its recent appreciation versus the U.S. dollar.4 We share this opinion, and expect the broad trade-weighted dollar to appreciate as U.S. growth rebounds in the back-half of the year.5 In our cross-sectional model, which adjusts spreads for credit rating and duration. Local Authorities and Foreign Agencies continue to look attractive compared to most U.S. corporate sectors. In contrast, the Sovereign and Supranational sectors appear expensive. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 38 basis points in July (before adjusting for the tax advantage). Munis have outperformed the Treasury benchmark by 186 bps year-to-date. The average Municipal / Treasury (M/T) yield ratio fell 2% in July, breaking below 85%. The average yield ratio remains extremely tight relative to its post-crisis trading range (Chart 6). There is more compensation available at the long-end of the muni curve than at the short-end (panel 2), and investors should continue to favor long maturities over short maturities on the Aaa Muni curve. Our early estimate, based on the recently released second quarter National Accounts data, shows that state & local government net borrowing probably moved higher in Q2 (panel 3), making the recent decline in yield ratios appear even more tenuous. The increase in net borrowing stems largely from a $21 billion drop in income tax revenues and a $20 billion decline in transfer receipts from the federal government. Income tax revenue should recover in the next two quarters,6 and we expect net borrowing will also start to decline. However, it is unlikely that net borrowing will fall by enough to justify current muni valuations. On July 6, the state House of Illinois overrode Governor Bruce Rauner's veto to finally pass a $36 billion budget. The move was sufficient for Moody's and S&P to both subsequently affirm the state's investment grade rating. The 10-year Illinois General Obligation bond yield declined 102 bps on the month, despite only a 1 bp drop in the 10-year Treasury yield. Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve bull steepened in July. The 2/10 slope steepened 3 bps and the 5/30 slope steepened 10 bps. We currently recommend two tactical trades designed to profit from movements in the Treasury curve. First, we have been recommending a short position in the July 2018 fed funds futures contract since July 11.7 From current levels, we calculate this trade will deliver an un-levered return of 28 bps if there are two hikes between now and then, and 53 bps if there are three hikes. Our second recommendation is a long position in the 5-year bullet versus a short position in a duration-matched 2/10 barbell, a trade designed to profit from a steepening of the 2/10 yield curve. It remains our view that inflation and inflation expectations, and not Fed tightening, are the main determinants of the slope of the yield curve. We expect the 2/10 slope to steepen as inflation rebounds during the next few months. Two weeks ago we published a Special Report 8 that explained our rationale for taking views on the slope of the curve using butterfly trades. It also explained our butterfly spread valuation model, and how we use that model to determine how much steepening/flattening is currently discounted in the yield curve. According to our model, the curve is priced for 9 bps of 2/10 steepening during the next six months (Chart 7). Our recommended butterfly trade will earn positive returns if the curve steepens by more than that. TIPS: Overweight Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 39 basis points in July. The 10-year TIPS breakeven inflation rate rose 9 bps on the month and, at 1.8%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. Core inflation has moved sharply lower since February, but the fact that our Phillips Curve model of core inflation has not rolled over makes us inclined to view the downtrend as transitory. Also, during the past few weeks we have seen some preliminary signs that inflation is on the cusp of rebounding. Year-over-year core PCE inflation ticked higher in June for the first time since January. The PCE diffusion index, which has a good track record capturing near-term swings in core PCE, moved sharply higher (Chart 8). The prices paid components of the ISM manufacturing and non-manufacturing surveys increased from 55 to 62 and from 52.1 to 52.7, respectively, in July. We expect stronger realized inflation will lead TIPS breakevens higher during the next few months. However, even in a scenario where core inflation fails to rebound, the downside in breakevens from current levels is limited. The reason is that if inflation remains very low, the Fed will most likely refrain from hiking rates in December. Such a dovish capitulation from the Fed would put upward pressure on breakevens at the long-end of the curve. We discussed this possible scenario in more detail in a recent report.9 ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 5 basis points in July, bringing year-to-date excess returns up to 59 bps. The index option-adjusted spread for Aaa-rated ABS held flat on the month, and remains well below its average pre-crisis level. The Federal Reserve released its Q2 Senior Loan Officer Survey last week. It showed that credit card lending standards moved back into "net tightening" territory after having eased the previous quarter (Chart 9). Auto loan lending standards tightened on net for the fifth consecutive quarter. Tightening lending standards are usually a response to deteriorating credit quality, and thus tend to correlate with higher losses and wider spreads. In that regard, net loss rates for auto loans continue to trend higher, and Moody's data show that the cumulative loss rate for prime auto loans originated in 2017 is worse than for any vintage since 2009, for loans with the same age. Conversely, the mild tightening in credit card lending standards has so far not translated into rising charge-offs (Chart 9), but the situation bears close monitoring. For now, we are content to remain overweight ABS given the attractive spread pick-up compared to other similarly risky sectors. However, we also recommend investors favor Aaa-rated credit cards over Aaa-rated auto loans, even though auto loans now once again offer an attractive spread differential, after adjusting for differences in duration and spread volatility (panel 3). Non-Agency CMBS: Underweight Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 39 basis points in July, bringing year-to-date excess returns up to 96 bps. The index option-adjusted spread for non-agency Aaa-rated CMBS tightened 4 bps on the month, and remains below its average pre-crisis level. The Fed's Q2 Senior Loan Officer Survey showed that lending standards for all classes of commercial real estate (CRE) loans tightened, on net, for the eighth consecutive quarter. The survey also reported that demand for CRE loans is on the decline (Chart 10). The combination of tighter lending standards and weak loan demand suggests that credit concerns continue to mount in the private CMBS space. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 11 basis points in July, bringing year-to-date excess returns up to 65 bps. The average option-adjusted spread for the Agency CMBS index held flat on the month but, at 49 bps, the sector continues to look attractive compared to other similarly risky alternatives.10 Not only does the sector offer attractive spreads, but the agency guarantee and the lower delinquency rate in multi-family loans compared to other CRE loans (panel 5) makes its risk/reward profile particularly appealing. Treasury Valuation Chart 11Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.62% (Chart 11). Our 3-factor version of the model, which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.63%. The U.S. PMI bounced back in July, after having trended lower for most of this year. The Chinese PMI also increased last month, while the Eurozone reading moderated somewhat from a very high level (panel 4). Overall, the Global PMI came in at 52.7 in July, up from 52.6 in June. Bullish sentiment toward the U.S. dollar has also fallen sharply in recent weeks (bottom panel). Bearish dollar sentiment in an environment of expanding global growth sends a very bond-bearish signal. It means that the entire world is participating in the global expansion and any increase in Treasury yields is less likely to be met with an influx of foreign buying. For further details on our Treasury models please refer to the U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com At the time of publication the 10-year Treasury yield was 2.26%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com Alex Wang, Research Analyst alexw@bcaresearch.com 1 Range calculated using monthly data, specifically the final day of each month. 2 Please see U.S. Bond Strategy Weekly Report, "Summer Snapback", dated July 11, 2017, available at usbs.bcaresearch.com 3 Ranges for junk spread and VIX calculated using monthly data, specifically the final day of each month. 4 Please see Emerging Markets Strategy Weekly Report, "The Case For A Major Top In EM", dated July 12, 2017, available at ems.bcaresearch.com 5 Mexico carries the largest weight in the Sovereign index, accounting for 23% of market cap. 6 Please see U.S. Bond Strategy Weekly Report, "Will The Fed Stick To Its Guns?", dated May 16, 2017, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy Weekly Report, "Summer Snapback", dated July 11, 2017, available at usbs.bcaresearch.com 8 Please see U.S. Bond Strategy Special Report, "Bullets, Barbells And Butterflies", dated July 25, 2017, available at usbs.bcaresearch.com 9 Please see U.S. Bond Strategy Weekly Report, "Three Scenarios For Treasury Yields In 2017", dated June 20, 2017, available at usbs.bcaresearch.com 10 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights Near record high levels for stocks are not an impediment to gains in the stock-to-bond ratio in the next 12 months. Minutes from June's FOMC meeting confirmed that policymakers agree that monetary policy should continue to normalize in the coming quarters. None of the main indicators that have provided some leading information in the past are warning of an equity bear market. Solid ISM and industrial production readings herald bullish profit growth in the second half the year. Treasury yields are headed higher in 2017, supporting our stocks over bond view. Within the U.S. bond market, we prefer short over long duration and investment-grade and high-yield bonds over high-quality debt; MBS will be hurt more than Treasuries as the Fed pares its balance sheet. Feature U.S. stocks will continue to reach all-time highs if inflation remains low, the economic backdrop fosters EPS growth and the Fed only gradually raises rates. We expect these conditions to stay in place in the second half of 2017 and into 2018, allowing stocks to outrun bonds. We note below that neither valuations nor technicals are flashing a red warning sign. Chart 1 shows that most of the time, even when equities are at record highs, valuations are above average (but not extreme) and the Fed is slowly removing accommodation, stocks can still rise. Moreover, none of the indicators that provided leading information in the past now warn of an equity bear market. Chart 1Macro Conditions Favorable For More Gains In Equities Macro Conditions Favorable For More Gains In Equities Macro Conditions Favorable For More Gains In Equities Chart 2Labor Market Strong But Wages Still Stagnant Labor Market Strong But Wages Still Stagnant Labor Market Strong But Wages Still Stagnant The June jobs report suggests that the environment of solid economic growth and still muted wage pressures remains in place, a positive backdrop for equity markets. The report showed that the economy added 222,000 jobs in June, well above the consensus forecast of 178,000. Prior months were also revised higher by 47,000 pushing the 3-month moving average up to 180,000 jobs per month. This is right in line with the 187,000 jobs per month averaged in 2016. Despite robust payroll gains, the unemployment rate actually ticked higher in June, from 4.3% to 4.4%, as previously sidelined workers were drawn back into the labor force. Meanwhile, wage growth continues to underwhelm, rising only 0.2% in June with the year-over-year growth rate holding steady at 2.5%. The deceleration in the 3 month change in average hourly earnings from 2.7% in December 2016 to 1.9% in June challenges the Fed's view on inflation (Chart 2). The recent moderation in wage growth is not yet severe enough to prevent the Fed from delivering one more rate hike before year-end. However, if the labor force participation rate continues to increase, and especially if this increase occurs alongside a rising unemployment rate, then the Fed's forecast of gradually accelerating wages will come into question. Fed Minutes: No Change To Our Base Case Minutes from June's FOMC meeting show that the debate among policymakers over monetary policy centers on the timing and pace of normalization in the coming quarters. The minutes did not provide any new insight about the Fed's plans to shrink its balance sheet. This will be done using caps on the monthly amount of principal repayments from the Fed's security holdings that will not be rolled over. These caps will rise over time on a pre-set path. The FOMC is still debating the timing of the start of this process. The FOMC was reasonably pleased with the tone of recent economic data, which support the view that GDP has bounced back from a soft patch in the first quarter. The June manufacturing and services ISM surveys, released since the FOMC meeting, undoubtedly reinforced policymakers' confidence in the underlying growth trajectory (see below for more details). The FOMC participants discussed at length the recent pullback in core measures of consumer price inflation. Most policymakers are willing for the time being to believe that inflation is driven primarily by temporary one-off factors. Others are worried that it will be more enduring. The moderation in three-month rates of change of prices this year was widespread across sectors of the CPI (i.e. it is not merely the result of one-offs). Inflation according to the Fed's favored measure, the core PCE price index, has also moderated this year although the disinflation has not been as broadly based as in the CPI (Chart 3). Much of the FOMC's debate focused on the relationship between labor market tightness and inflation. The doves want to see inflation rise closer to the 2% target before tightening even more. The hawks worry that the relationship could be non-linear, which means that a further undershoot of unemployment below estimates of full employment could suddenly generate a surge in inflation. At a minimum, an undershoot could boost risks to financial stability by promoting excess risk-taking in the financial markets. The minutes reveal that the worries about the impact of easing financial conditions on financial stability have intensified since the start of the year. Inflation forecasting has been particularly tricky since the Great Recession for both the Fed and other economic prognosticators. Admittedly, it is difficult to explain the sudden and broadly-based inflation deceleration, even in sectors that have nothing to do with oil prices, shifts in the currency or wage growth. That said, the model shown in the top panel of Chart 4 suggests that core CPI inflation will edge higher in the coming months. This reflects the acceleration in ECI wage growth (feeding into higher core services inflation) and in core goods inflation (reflecting rising import prices), which more than offset the slight moderation in our projection for shelter inflation. Chart 3Inflation Readings Must##BR##Improve In Next Few Months Inflation Readings Must Improve In Next Few Months Inflation Readings Must Improve In Next Few Months Chart 4Core CPI Should Edge Higher##BR##In Coming Months Core CPI Should Edge Higher In Coming Months Core CPI Should Edge Higher In Coming Months Bottom Line: The minutes did not change our base case outlook; the FOMC will announce in September that it will begin to shrink the Fed's balance sheet shortly thereafter. The next rate hike will occur in December. Nonetheless, this forecast hangs importantly on the assumption that core inflation edges higher in the coming months. We think it will, but uncertainty is high. Monitoring The Bear Market Barometer The FOMC's seeming determination to stick with the current tightening timetable raises question marks over the equity market, especially given elevated valuations. Chart 5Equity Bear Market Indicators Equity Bear Market Indicators Equity Bear Market Indicators BCA's Chief Economist, Martin Barnes, highlighted the best "equity bear market" indicators to watch in a 2014 Special Report1. He noted that no two bear markets are the same and that there are no indicators that have reliably heralded bear phases. Nonetheless, there are some common elements. The safest time to invest in the market is when monetary conditions are favorable, there are no signs of a looming economic downturn, extreme overvaluation is not present and technical indicators are not flashing red. Some indicators related to each of these fundamental factors are shown in Chart 5: Monetary Conditions: The yield curve is flat by historical standards, but it is far from inverted. Moreover, real short-term interest rates are usually substantially higher than today, and above 2%, when bear markets commence. Excess liquidity, which we define as M2 growth less nominal GDP growth, is also well above the zero line, a threshold that in the past has warned of a downturn in stock prices. Valuation: Our composite valuation indicator is still shy of the +1 standard deviation level that defines over-valued. However, this is due to the components that compare equity prices to bond yields. The other three components of the equity indicator, which are unrelated to bond yields, suggest that stock valuation is stretched. Economic Outlook: Economic data, such as the leading economic indicator and ISM, have been unreliable bear market signals. We do not see anything that indicates that a recession is on the horizon. U.S. growth will remain above-trend in the second half of the year based on its relationship with financial conditions. Technical Conditions: Sentiment is elevated, which is bearish from a contrary perspective. However, breadth, the deviation from the 40- week moving average and our composite technical indicator, all are not flashing red. Earnings: Trends in earnings and margins did not provide any additional reliable signals for timing equity market downturns in the past. Still, it is a bad sign when EPS growth tops out. This is often preceded by a peak in industrial production growth. We expect EPS growth to continue to accelerate for at least a few more months, but we are closely watching industrial production. Bottom Line: The equity market is vulnerable to unforeseen shocks given stretched valuation. Nonetheless, none of the indicators that have provided leading information in the past warn of an equity bear market. ISM Above 50 Supports 2H Profit Outlook The elevated level of ISM sets the stage for EPS growth to gather speed in the second half of 2017. Industrial production is a good proxy for sales of S&P 500 companies (Chart 6). A rollover in the 12-month change in IP would challenge our view. However, strong readings on the ISM, which tracks IP, suggest that IP should accelerate in the next six months (Chart 6, panel 1). Chart 6Solid Backdrop For Earnings And Sales Solid Backdrop For Earnings And Sales Solid Backdrop For Earnings And Sales At 57.8 in June, the ISM has rebounded from the recent low of 47.9 in 2015. Investors wonder if it will roll over again or simply fluctuate at a high level. The leading components of ISM, including the new orders index and the new orders-to-inventory ratio, indicate that the ISM will remain above 50 in the months ahead (Chart 7). Moreover, the new export orders component of the ISM has also surged. The implication is that foreign demand (rather than domestic consumer or business spending) is leading the U.S. manufacturing sector. In fact, the 3- and 12-month change in the industrial production indices in advanced economies outside the U.S. have outpaced domestic growth (Chart 8). Chart 7IP Poised To Accelerate IP Poised To Accelerate IP Poised To Accelerate Chart 8U.S. IP Lagging Other Developed Markets U.S. IP Lagging Other Developed Markets U.S. IP Lagging Other Developed Markets Bottom Line: Firm readings on ISM are an indication that our bullish profit story for 2017 remains intact. Stay overweight stocks versus bonds. Inflection Point The increase in Treasury yields since late June indicates that growth expectations had become overly pessimistic. Our assessment is that U.S. growth will remain above trend for the rest of 2017. The implication for investors is that Treasury bond yields will move higher, the yield curve will bear-steepen, and that credit will outperform Treasuries in the second half of 2017. Moreover, we expect MBSs to underperform. According to our U.S. Bond Strategy service2, Treasury yields are poised to follow the economic surprise index higher in the coming months. Extreme net long positioning in the futures market supports the view. The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.52%. Our 3-factor version of the model, which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.45% (Chart 9). Investors should continue to position for a steeper curve by favoring the 5-year bullet versus a duration-matched 2/10 barbell. Small positive excess returns, consistent with carry, remain the most likely scenario for investment- grade credit, where we recommend an overweight. We do not see the potential for much spread tightening from current levels. However, a large spread widening would be equally unlikely given the favorable backdrop of steady growth and muted inflation. We recommend an overweight in the high-yield market. We expect the decline in the 12-month trailing speculative default rate to continue for the rest of the year, aided by a moderation in energy sector defaults (Chart 10, bottom panel). This means that the current compensation offered by junk spreads in excess of expected default losses stands at 221 bps, in line with its historical average (Chart 10, panel 3). In last week's Weekly Report3 our U.S. Bond Strategy team showed that a default-adjusted spread of 221 bps is consistent with excess returns close to 150 bps during the next 12 months. Chart 9Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models Chart 10High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview Our Energy Sector Strategy team stated in a Weekly Report4 last week that our base case of $50-$60/bbl WTI crude oil prices by the end of 2017 should keep high-yield energy spreads contained. We remain underweight MBSs. Nominal MBS spreads are already very tight compared with previous levels, and they appear even tighter relative to trends in net issuance. While refinancing activity will remain depressed, we see potential for option-adjusted spreads to follow net issuance higher, even as the compensation for prepayment risk (option cost) remains low. The Fed's exit from the MBS market, which could occur as early as September, represents an additional upside risk for spreads. Bottom Line: Rates have bounced up after undershooting between March and the end of June. Loftier inflation readings are needed to sustain the bounce. Higher rates in the rest of 2017 support our stocks-over-bond stance. Within the U.S. bond market, we favor short duration over long, and credit over high-quality. MBSs will be hurt more than Treasurys as the Fed begins to shrink its balance sheet. John Canally, CFA, Senior Vice President U.S. Investment Strategy johnc@bcaresearch.com Mark McClellan, Senior Vice President The Bank Credit Analyst markm@bcaresearch.com 1 Please see BCA Special Report "Timing The Next Equity Bear Market, " dated January 24, 2014, available at bca.bcaresearch.com. 2 Please see U.S. Bond Strategy Portfolio Allocation Summary, "Inflection Point", dated July 5, 2017, available at usbs.bcaresearch.com. 3 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com. 4 Please see Energy Sector Strategy Weekly Report, "HY Debt Update: Offshore Drilling & Transportation Getting Left Behind", dated July 5, 2017, available at nrg.bcaresearch.com.

Highlights Chart 1Too Pessimistic On Growth Too Pessimistic On Growth Too Pessimistic On Growth Treasury yields bounced sharply last week and the yield curve steepened. As a result the Bloomberg Barclays Treasury index posted a negative return in June, only the second month of negative Treasury returns so far in 2017. Last week's increase in yields could signal that growth expectations have finally become overly pessimistic. Our U.S. Investment Strategy service has calculated that after the U.S. Economic Surprise Index rises above 40, its average peak to trough decline lasts 90 days. Given that the surprise index peaked above 40 in mid-March, a bottoming-out in the coming weeks would be right on schedule (Chart 1). Net speculative positioning in the futures market has also capitulated, swinging sharply from net short to net long. In recent years, extreme net long positioning has led to higher Treasury yields during the following three months (bottom panel). Our assessment is that U.S. growth will remain above trend for the remainder of the year, and the Treasury curve will continue to bear-steepen as the economic data start to outperform downbeat expectations. Stay at below-benchmark duration, in curve steepeners, overweight spread product versus Treasuries, and overweight TIPS versus nominals. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 50 basis points in June. The index option-adjusted spread tightened 4 bps to end the month at 109 bps. Though below its historical mean, the investment grade spread is actually somewhat elevated compared to the early stages of prior Fed tightening cycles (Chart 2). We calculate that in the early stages of the past two tightening cycles (February 1994 to July 1994 & June 2004 to December 2005), the index option-adjusted spread averaged 90 bps and traded in a range between 66 bps and 107 bps. While spreads are currently more attractive than is typical for this stage of the cycle, there is good reason for investors to demand some extra risk premium. In a recent report1 we observed that non-financial corporate debt as a percent of GDP is already as high as it was during the past two recessions. Further, the majority of this debt has been issued to finance direct payments to shareholders (dividends & buybacks) as opposed to capital investment. This unfavorable shift in corporate capital structures means that bond investors should demand somewhat greater compensation. All in all, we do not see potential for much spread tightening from current levels. However, a large spread widening would be equally unlikely given the favorable back-drop of steady growth and muted inflation. Small positive excess returns, consistent with carry, remains the most likely scenario. Energy debt underperformed duration-matched Treasuries by 12 bps in June. The sector still looks cheap after adjusting for credit rating and duration (Table 3), and our commodity strategists remain bullish on oil. Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Inflection Point? Inflection Point? Table 3BCorporate Sector Risk Vs. Reward* Inflection Point? Inflection Point? High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 35 basis points in June. The index option-adjusted spread widened 1 bp to end the month at 364 bps, 20 bps above its 2017 low. Energy sector spreads widened sharply in June, alongside falling oil prices, once again de-coupling from the overall index spread (Chart 3). Junk-rated energy credits underperformed the duration-equivalent Treasury index by 190 bps in June, while the High-Yield index excluding energy outperformed by 70 bps. In a report published today,2 our Energy Sector Strategy service takes a detailed look at credit risk among high-yield energy issuers, concluding that while the worst of the energy bankruptcy cycle is behind us, $23 billion of high-yield energy debt remains in distress. 91% of that distressed debt is in the Exploration & Production and Offshore Drilling & Transportation sectors. The continued moderation in energy sector defaults will ensure that the overall speculative grade default rate trends lower for the rest of the year, probably settling below 3% (bottom panel). The decline in defaults means that the current compensation offered by junk spreads in excess of expected default losses stands at 221 bps, right in line with its historical average (panel 3). In last week's report,3 we showed that a default-adjusted spread of 221 bps is consistent with excess returns close to 150 bps during the next 12 months. MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 20 basis points in June, dragging year-to-date excess returns down to -20 bps. The conventional 30-year MBS yield rose 11 bps on the month, driven by a 7 bps increase in the rate component and a 6 bps widening of the option-adjusted spread (OAS). This was partially offset by a 2 bps decline in the compensation for prepayment risk (option cost). In last week's report,4 we examined the risk/reward trade-off in different Aaa-rated spread products. We found that despite some recent widening in MBS OAS, you still need to move into 4% coupons or higher to find competitive spreads relative to Aaa-rated corporates, consumer ABS, agency CMBS and non-agency CMBS. Further, MBS OAS are still too tight compared to the trend in net issuance (Chart 4), and even though depressed refi activity will continue to hold down the option cost component of spreads, it is unlikely that a lower option cost will be able to completely offset wider OAS during the next 12 months. The Fed released more details about its balance sheet run-off plan at the June FOMC meeting. We now know that the Fed will start by allowing only $4 billion of MBS per month to run off its balance sheet, but this cap will increase by $4 billion every 3 months until it reaches $20 billion per month. This means that even if the Fed starts to wind down its balance sheet following the September meeting, which is our base case expectation, then it will still be some time before a significant amount of extra supply shifts into the private market. Government-Related: Underweight Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 21 basis points in June, bringing year-to-date excess returns up to +107 bps. Sovereigns and Local Authorities outperformed the Treasury benchmark by 65 bps and 73 bps, respectively. The low-beta Supranational and Domestic Agency sectors outperformed by 2 bps and 10 bps, respectively. The Foreign Agency sector underperformed duration-matched Treasuries by 4 bps, alongside the dip in oil prices. A weakening U.S. dollar has led to the outperformance of USD-denominated sovereign debt so far this year. Year-to-date, the Sovereign index has outperformed the duration-equivalent Treasury index by 300 bps. This is better than the equivalently-rated Baa U.S. Corporate index, which has outperformed by 195 bps year-to-date. However, there are already signs that the trade-weighted dollar is starting to moderate its downtrend (Chart 5), and we expect the trade-weighted dollar will strengthen as the economic data surprise to the upside in the back half of the year, as discussed on the first page of this report. Granted, the Mexican peso continues to strengthen versus the dollar (panel 3) and this currency pair is particularly important since Mexico is the largest issuer in the Sovereign index. On the heels of its recent outperformance, the Sovereign sector once again looks expensive compared to U.S. corporate sectors, after adjusting for credit rating and duration. Meanwhile, the Local Authority and Foreign Agency sectors continue to look cheap. Supranationals and Domestic Agencies offer very little additional compensation relative to Treasuries, and as we discussed last week,5 there are better options available for investors in need of high-quality spread product. Municipal Bonds: Underweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds underperformed the duration-equivalent Treasury index by 18 basis points in June (before adjusting for the tax advantage). Last month we observed that Municipal / Treasury (M/T) yield ratios had become very tight, and we advised reducing municipal bond exposure to underweight. The average M/T yield ratio ticked higher in June, but at 85%, it remains more than one standard deviation below its post-crisis average (Chart 6). There is more compensation available at the long-end of the muni curve than at the short-end (panel 2), and investors should continue to favor long maturities over short maturities on the Aaa Muni curve. The National Association of State Budget Officers recently released its Fiscal Survey of the States and it showed that overall general fund expenditures are expected to increase by only 1% in the 2018 fiscal year, the slowest rate of growth since 2009/10. Meanwhile, 23 states have already enacted mid-year budget cuts in 2017. Budget cutting measures are clearly a response to disappointing tax revenues, which should bounce back somewhat in fiscal year 2018.6 This will help reduce net borrowing, though probably not by enough to justify current municipal bond valuations (panel 3). The state of Illinois avoided a ratings downgrade to junk this week, as the State House of Representatives voted to approve an income tax increase. This measure will keep the rating agencies at bay for now, but a downgrade is still possible in the coming months if the state fails to pass a budget for fiscal year 2018. Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve bull-flattened for most of June, before suddenly reversing course and bear-steepening late in the month. The 2/10 slope flattened 15 basis points between the end of May and June 26, and then steepened 15 bps between June 26 and the end of the month. All told, the 2/10 slope was unchanged in June, while the 5/30 slope flattened 17 bps. The abrupt transition from bull-flattening to bear-steepening was prompted by comments from European Central Bank (ECB) President Mario Draghi that suggested a much more hawkish bias from the ECB. Higher rate expectations in the rest of the world should put downward pressure on the U.S. dollar, and historically, bearish sentiment toward the U.S. dollar has led to a steeper U.S. yield curve (Chart 7, bottom panel). This correlation has not held up so far this year, and we suspect this is because a weaker dollar has not translated into higher U.S. inflation and inflation expectations, as it usually does. We have previously made the case that inflation and inflation expectations, and not Fed tightening, are the main determinants of the slope of the yield curve (panel 4).7 As such, we attribute the bulk of this year's curve flattening to disappointing core inflation which has dragged TIPS breakevens lower. This should reverse in the coming months.8 Investors should continue to position for a steeper curve by favoring the 5-year bullet versus a duration-matched 2/10 barbell. TIPS: Overweight Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS underperformed the duration-equivalent nominal Treasury index by 86 basis points in June. The 10-year TIPS breakeven rate fell 8 bps on the month and, at 1.75%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. In a recent report9 we outlined three possible scenarios for Treasury yields between now and the end of the year based on the interaction between incoming inflation data and Fed policy. In our base case scenario inflation will start to rebound in the coming months, heeding the message from our Phillips Curve model (Chart 8), leading to wider TIPS breakevens and keeping the Fed on its current tightening path. Even if realized inflation remains depressed, the next most likely scenario is that the Fed will capitulate later this year and adopt a shallower expected rate hike path. Such a dovish reaction from the Fed would lend support to long-maturity breakeven wideners, even though real yields would decline. The least likely scenario, in our view, is one where realized inflation remains low but the Fed sticks to its hawkish rhetoric. This is also the scenario that would lead to the most downside in the cost of inflation protection. May PCE inflation data were released last Friday, with year-over-year core PCE decelerating from 1.50% to 1.39%, and trimmed mean PCE decelerating from 1.70% to 1.66% (panel 4). One bright spot is that our PCE Diffusion Index swung sharply into positive territory. Historically, this index has a strong track record signaling turning points in core inflation (bottom panel). ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 2 basis points in June, bringing year-to-date excess returns up to +54 bps. The index option-adjusted spread for Aaa-rated ABS tightened 2 bps on the month, and remains well below its average pre-crisis level. Despite low spreads relative to history, in a recent report10 we showed that Aaa-rated ABS appear quite attractive compared to other Aaa-rated spread product. Specifically, Aaa consumer ABS offer greater compensation per unit of duration than Agency bonds, agency MBS and Aaa Credit. They offer similar compensation per unit of duration to Agency CMBS, but less than non-Agency Aaa CMBS. Within consumer ABS, auto loan-backed securitizations offer slightly greater compensation than the credit card-backed variety (Chart 9). However, we still prefer credit card ABS over auto loan ABS. While credit card charge-offs remain historically low, auto net loss rates are rising. Auto lending standards also moved deeper into "net tightening" territory in the first quarter, according to the Fed's Senior Loan Officer Survey, while credit card lending standards dipped back into "net easing" territory (bottom panel). We continue to recommend that investors favor Aaa-rated credit cards over Aaa-rated auto loans within an overall overweight allocation to consumer ABS. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 5 basis points in June, bringing year-to-date excess returns up to +57 bps. The index option-adjusted spread for non-agency Aaa-rated CMBS tightened 1 bp on the month, and remains below its average pre-crisis level (Chart 10). In last week's report,11 we showed that non-agency CMBS offer by far the most compensation per unit of duration of any Aaa-rated spread sector. However, we are concerned that non-agency CMBS spreads will widen on a 6-12 month horizon. Commercial real estate lending standards are tightening and property prices are decelerating. Both of these developments tend to correlate with wider spreads. Despite lower spreads, we are much more comfortable in the Agency CMBS market. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 4 basis points in June, bringing year-to-date excess returns up to +54 bps. Agency CMBS offer somewhat lower spreads than their non-agency counterparts, but this sector should be more insulated from spread widening in the months ahead. Not only do these securities benefit from agency backing, but they also mostly comprise multi-family loans. Multi-family property prices have been stronger than those in the retail and office sectors, and delinquencies have been lower (bottom 2 panels). Treasury Valuation Chart 11Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.52% (Chart 11). Our 3-factor version of the model, which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.45%. The lower fair value results from the large spike in the uncertainty index last November, which has only been partially unwound. The U.S. PMI has dipped lower in recent months, but remains firmly entrenched above the 50 boom/bust line. Meanwhile, the Eurozone PMI continues to surge ahead. China's PMI sent a worrying signal when it dipped below 50 in May, but it bounced back to 50.4 last month (bottom panel). Overall, the Global PMI came in at 52.6 in June, no change from the prior month. For further details on our Treasury models please refer to the U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com At the time of publication the 10-year Treasury yield was 2.35%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com Alex Wang, Research Analyst alexw@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "Low Inflation And Rising Debt", dated June 13, 2017, available at usbs.bcaresearch.com 2 Please see Energy Sector Strategy Weekly Report, "HY Debt Update: Offshore Drilling & Transportation Getting Left Behind", dated July 5, 2017, available at nrg.bcaresearch.com 3 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com 5 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com 6 For further details please see U.S. Bond Strategy Weekly Report, "Will The Fed Stick To Its Guns?", dated May 16, 2017, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy Weekly Report, "The Yield Curve On A Cyclical Horizon", dated March 21, 2017, available at usbs.bcaresearch.com 8 Please see U.S. Bond Strategy Weekly Report, "Three Scenarios For Treasury Yields In 2017", dated June 20, 2017, available at usbs.bcaresearch.com 9 Please see U.S. Bond Strategy Weekly Report, "Three Scenarios For Treasury Yields In 2017", dated June 20, 2017, available at usbs.bcaresearch.com 10 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com 11 Please see U.S. Bond Strategy Weekly Report, "Risk Rally Extended", dated June 27, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon)

Highlights Fed Policy Loop: Low inflation is preventing rate hike expectations from being revised higher, prolonging the current rally in spread product. We expect rate hike expectations to move up as inflation recovers, eventually leading to a correction in spread product. Such a correction will prove fleeting as long as inflation stays below target. High-Yield: High-yield valuation is consistent with its historical average, after accounting for expected default losses. Current valuation levels should translate into excess returns of just over 200 bps during the next 12 months. Aaa Spread Product: Non-agency CMBS offer the most spread pick-up of any Aaa-rated sector. However, we prefer to focus our Aaa-rated spread product allocation in Agency CMBS and credit card backed consumer ABS. Feature Chart 1The Fed Policy Loop In Action The Fed Policy Loop In Action The Fed Policy Loop In Action One of this publication's main themes during the past few years has been the Fed Policy Loop.1 In essence, the Loop describes the feedback mechanism between monetary policy and financial markets, a relationship that results from both investors' sensitivity to the Fed's policy stance and the Fed's reliance on financial conditions as a predictor of economic growth. In practice, the Loop works as follows: Easier Fed policy causes spread product to outperform Treasuries. Tighter credit spreads lead to easier financial conditions, which the Fed interprets as a sign that economic growth will accelerate. An improved economic outlook causes the Fed to step up the pace of tightening. Tighter Fed policy causes spread product to underperform Treasuries. Wider credit spreads lead to tighter financial conditions, which the Fed interprets as a sign that economic growth will moderate. A worse economic outlook causes the Fed to slow its expected pace of tightening. Rinse, repeat. Chart 2 Chart 2 provides a graphical description of the Loop, and its most recent iteration can be seen in Chart 1 above. Chart 1 shows that corporate bonds outperformed Treasuries leading up to the March rate hike, but then rate expectations rose too far. In mid-March the market was discounting a fed funds rate of 1.86% by the end of 2018. These overly stringent rate hike expectations caused corporate bonds to underperform, and this underperformance led rate hike expectations to be revised lower. The market now expects a fed funds rate of only 1.47% by the end of 2018, and these depressed rate expectations have fueled the rally in corporate bonds that started in mid-April. Normally at this stage of the Fed Policy Loop we would expect rate hike expectations to move higher until they eventually prompt a correction in corporate spreads. However, extremely disappointing core inflation prints during the past three months have caused the market to keep its rate hike expectations depressed. This has extended the most recent rally in spread product. This is why we have consistently pointed to core inflation and the cost of inflation protection embedded in long-maturity bond yields as the main factors to watch to determine how much life is left in the corporate bond trade. As long as inflation stays below target, the Fed absolutely needs it to rise. It will therefore be quick to respond to any tightening of financial conditions/widening of credit spreads. Table 1 shows average monthly excess returns for investment grade corporate bonds relative to duration-matched Treasuries. These returns are split into buckets based on the reading from the St. Louis Fed's Price Pressures Measure (PPM). The PPM is a composite of 104 economic indicators designed to measure the probability that inflation will exceed 2.5% during the next 12 months. As can be seen, average monthly excess returns are strongest when inflation pressures are low, but they gradually decline as inflation heats up and the Fed's reaction function becomes less supportive for markets. At present, the PPM gives a reading of only 4.8%. Table 1Investment Grade Corporate Bond Excess Returns* Under Different ##br## Ranges Of Price Pressures Measure** (January 1990 To Present) Risk Rally Extended Risk Rally Extended Similarly, Table 2 shows that it is difficult to get a long-lasting correction in an environment with low inflation pressures and a responsive Fed. This table shows the results of a "buy the dips" trading strategy where if the average junk spread widens by 20 basis points we buy the junk index versus duration-matched Treasuries and hold it for a period of 1, 2 or 3 months. Just as in Table 1, this strategy works well when inflation pressures are muted, but starts to fail as inflation ramps up. Table 2High-Yield Corporate Bond Returns* Achieved By Holding The Junk Index ##br## Following A 20 BPs Widening In High-Yield Corporate OAS** Under Different Ranges Of ##br## The St. Louis Fed Price Pressures Measure*** (February 1994 To Present) Risk Rally Extended Risk Rally Extended Beatings Will Continue Until Morale Improves So when will the Fed staunch the current rally? That depends on how quickly inflation rebounds,2 and also on how much emphasis Fed policymakers place on financial conditions versus the actual inflation data. At the moment, most indexes are sending the message that financial conditions are easier than average and that the Fed should continue to tighten (Chart 3). However, as was noted above, inflation gauges are sending the opposite signal (Chart 3, panel 2). For now, the Fed is downplaying low inflation as transitory. It decided to leave its median projected rate hike path unchanged at the June FOMC meeting. But the Fed's refusal to capitulate in the face of weaker inflation has caused the yield curve to flatten, the cost of inflation protection to plummet (Chart 3, bottom panel) and investors to grow increasingly concerned about a policy mistake (Chart 4). Chart 3Financial Conditions Are Supportive Financial Conditions Are Supportive Financial Conditions Are Supportive Chart 4Should The Fed Keep Tightening? Should The Fed Keep Tightening? Should The Fed Keep Tightening? This brings up an interesting flaw in the financial conditions approach to policymaking. Most indexes of financial conditions are at least partially driven by long-maturity Treasury yields (lower yields = easier financial conditions, and vice-versa). This makes some sense. Lower yields do in fact indicate that the financing back-drop is more supportive and tend to translate into higher growth in the future (Chart 5). Chart 5Financial Conditions Lead Economic Growth Financial Conditions Lead Economic Growth Financial Conditions Lead Economic Growth However, what if lower long-maturity Treasury yields are the result of excessively tight Fed policy? This would appear to be the case at the moment. Investors are revising their long-run inflation forecasts lower on the view that the Fed is not doing enough to allow prices to rise. In such a situation it would be incorrect to interpret lower Treasury yields as a signal that policy needs to tighten further. On the contrary, tighter policy would only exacerbate the downtrend in yields. For this reason we do not include the level of yields in the financial conditions component of our Fed Monitor (Chart 3, top panel). As a result, this financial conditions indicator is not as deep in "easing territory" as most other indicators. However, it is still above the zero line, suggesting that policy should be biased tighter at the margin. Bottom Line: Low inflation is preventing rate hike expectations from being revised higher, prolonging the current rally in spread product. We expect rate hike expectations to move up as inflation recovers, eventually leading to a correction in spread product. Such a correction will prove fleeting as long as inflation remains below the Fed's target. The key risk is that inflation stays low but the Fed continues to focus exclusively on "easy" readings from financial conditions indexes, and proceeds on its current tightening path. In that scenario cries of "policy mistake" will grow louder and spread product will sell off, converging with lower rate hike expectations. We view this scenario as a low-probability tail risk. Junk Valuation Update At 378 bps, the average spread on the Bloomberg Barclays High-Yield index is only 55 bps above its post-crisis low, but still more than 100 bps above the level where it tends to settle in the late stages of the economic cycle when the Fed is tightening policy (Chart 6, top panel). Higher debt levels than are typical for this stage of the cycle suggest that somewhat wider spreads are justified,3 but the idea that junk spreads are extremely tight compared to history does not hold up to scrutiny. Chart 6High-Yield Default-Adjusted Spread High-Yield Default-Adjusted Spread High-Yield Default-Adjusted Spread Our preferred measure of junk valuation, the default-adjusted high-yield spread, paints an even rosier picture. The second panel of Chart 6 shows an ex-post measure of the default-adjusted spread (the option-adjusted spread of the high-yield index less actual default losses over the subsequent 12 months). The most recent reading from this indicator is based on our forecast of default losses for the next 12 months, and is shown as a dashed line. The message from the default-adjusted spread is that, assuming our default loss forecast is correct, junk bonds currently offer compensation for default risk that is in line with the historical average. That level of compensation would be consistent with an excess return of just over 200 bps during the next 12 months (Chart 6, panel 3), and is contingent on the speculative grade default rate falling to 2.68%, in line with Moody's baseline forecast (Chart 6, bottom panel). We expect a decline in the default rate to materialize in the coming months as commodity sector defaults continue to work their way out of the data. Moody's did not record any commodity-related defaults in May, the first month this has occurred since January 2015. The risk going forward is that defaults start to emerge in the increasingly stressed retail sector. So far, Moody's has recorded two retail defaults this year. However, more are probably on the way. This will be especially true if inflationary pressures start to mount and the Fed tightens policy, giving banks less incentive to extend credit. We will be monitoring the situation in retail closely going forward. Bottom Line: High-yield valuation is consistent with its historical average, after accounting for expected default losses. Current valuation levels should translate into excess returns of just over 200 bps during the next 12 months. Aaa Roundup As can be inferred from the previous two sections, we are still reasonably comfortable taking credit risk in U.S. bond portfolios. However, this week we also look at the compensation offered by Aaa-rated spread product. For investors who desire some Aaa-rated allocation outside of the Treasury market, Chart 7 provides a snapshot of where the most additional spread is available. Chart 7 The first thing that jumps out is that Agency bonds offer very little spread compared to other Aaa-rated instruments. Agency residential mortgage-backed securities also offer relatively little compensation, unless one is willing to extend into premium coupons (4% and above). Agency CMBS, auto ABS and credit card ABS all offer more spread than Aaa-rated corporate bonds. Non-agency CMBS offer much more attractive spreads than the other Aaa sectors, but we see potential for capital losses in that segment, as is discussed below. Agency MBS Only agency MBS carrying coupons of 4% or above offer interesting compensation relative to other Aaa-rated sectors, and even there we see potential for spread widening in the coming months. Nominal MBS spreads are already very tight compared to history (Chart 8) and appear even tighter relative to trends in net issuance (Chart 8, panel 2). While refinancing activity will likely stay depressed (Chart 8, panel 3), we see potential for option-adjusted spreads to follow net issuance higher, even as the compensation for prepayment risk (option cost) remains low. A similar scenario played out in 2007 (Chart 8, bottom panel). The Fed's exit from the MBS market, which could occur as early as September, represents an additional upside risk for spreads. Chart 8MBS Spreads Biased Wider MBS Spreads Biased Wider MBS Spreads Biased Wider Chart 9Avoid Non-Agency CMBS Avoid Non-Agency CMBS Avoid Non-Agency CMBS CMBS As noted above, non-agency CMBS look very attractive compared to other Aaa-rated spread products. But we see potential for spread widening in this sector. Commercial real estate lending standards are tightening and property prices are decelerating, both should pressure non-agency CMBS spreads wider (Chart 9). Agency CMBS offer somewhat lower spreads than their non-agency counterparts. But this sector should be more insulated from spread widening. For one thing, Agency CMBS are mostly backed by multi-family loans. Multi-family property prices have been stronger than those in the retail or office segments (Chart 9, panel 3), and multi-family properties have also experienced much lower delinquencies (Chart 9, bottom panel). Consumer ABS Chart 10Credit Cards Greater Than Autos Credit Cards > Autos Credit Cards > Autos While Chart 7 shows that Aaa-rated auto ABS offer a slight spread advantage over Aaa-rated credit card ABS, we are inclined to view credit card ABS more favorably. Rising auto loan net loss rates pose a risk for auto ABS spreads, while credit card charge-offs remain historically low (Chart 10). Auto lending standards have also moved deep into "net tightening" territory, while credit card lending standards have dipped back into "net easing" territory. The small extra compensation available in auto ABS relative to credit card ABS does not seem to be worth the extra risk. Bottom Line: Non-agency CMBS offer the most spread pick-up of any Aaa-rated sector. However, we view the risk of a further widening in non-agency CMBS spreads as substantial. We prefer to focus our Aaa-rated spread product exposure in Agency CMBS and credit card backed consumer ABS. Both sectors offer reasonably attractive spreads, and should remain insulated from capital losses going forward. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "Caught In A Loop", dated September 29, 2015, available at usbs.bcaresearch.com 2 Our view is that core inflation will rebound fairly quickly. For further details please see U.S. Bond Strategy Weekly Report, "Three Scenarios For Treasury Yields In 2017", dated June 20, 2017, available at usbs.bcaresearch.com 3 For further details please see U.S. Bond Strategy Weekly Report, "Low Inflation And Rising Debt", dated June 13, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights Chart 1Something's Got To Give Something's Got To Give Something's Got To Give Last Friday's disappointing employment report reinforced the bond market's recent strength. The 10-year Treasury yield reached a new 2017 low of 2.15%, the 10-year TIPS breakeven inflation rate broke below 1.8% and the overnight index swap curve is now priced for only 47 bps of rate hikes during the next 12 months. Increasingly, the bond market is discounting two different future states of the world that cannot possibly coexist. Decelerating wage growth has caused the market to expect fewer Fed rate hikes, while concurrently, the cost of long-maturity inflation protection has fallen and the yield curve has flattened (Chart 1). This means the market expects that poor wage growth and inflation will cause the Fed to back away from its expected pace of two more rate hikes this year, but also that this relent will not be sufficient to prompt a recovery in economic growth or inflation. This dichotomy cannot exist for long. Either wage growth and inflation will bounce back in the second half of the year allowing the Fed to lift rates twice more in 2017 (our base case expectation), or inflation will continue to disappoint in which case the Fed will slow its pace of hikes. In both cases long-maturity Treasury yields should head higher, led by an increasing cost of inflation compensation. Stay at below benchmark duration. Feature Investment Grade: Overweight Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 37 basis points in May. The index option-adjusted spread tightened 3 bps on the month and, at 113 bps, it remains well below its historical average (134 bps). Limited inflationary pressure will keep monetary policy accommodative enough to ensure excess returns consistent with carry. However, corporate spreads have already discounted a substantial improvement in leverage (Chart 2) and we do not see much potential for spread tightening from current levels. BEA data show that EBITD contracted in Q1, causing the annual growth rate to tick back below zero (panel 4). Meanwhile, gross issuance has been strong so far this year, suggesting that leverage will show an uptick in Q1 when the Flow of Funds data are released later this week. This aligns with our observation that, historically, net leverage - defined as total debt less cash as a percent of trailing EBITD - has never declined unless prompted by a recession. In other words, the corporate sector never voluntarily undertakes deleveraging, it only starts to pay down debt when forced by a severe economic contraction. For now, rising leverage will limit the amount of spread tightening, but shouldn't lead to negative excess returns. That will only occur when inflationary pressures are more pronounced and the Fed steps up the pace of tightening - probably sometime next year. Energy related sectors still appear cheap on our model (Table 3), and have outperformed the overall corporate index this year even though the oil price has fallen. Remain overweight. Chart Chart High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 48 basis points in May. The index option-adjusted spread tightened 8 bps on the month and, at 362 bps, it is currently 18 bps above its 2017 low. While the average spread on the junk index is a mere 38 bps above its post-crisis low, our estimate of the default-adjusted high-yield spread is 204 bps, only slightly below its historical average (Chart 3). Assuming our forecast for default losses is correct, a default-adjusted spread in this range has historically coincided with positive 12-month excess returns to high-yield bonds 74% of the time, with an average excess return of 82 bps. Our estimate of 12-month forward default losses is calculated using Moody's baseline assumption for the speculative grade default rate, which stands at 2.96%. We also incorporate an expected recovery rate of 47%. This expectation for a continued decline in the default rate squares with trends in corporate lending standards (which are once again easing), industrial production (which is accelerating) and job cut announcements (which are trending lower). Weak first quarter profit growth will be a headwind if it persists, but we expect it will recover alongside the broader economy in Q2. Overall, with muted inflationary pressures, an improving default back-drop and still moderate valuations, we think junk bonds will deliver small positive excess returns during the next 12 months. Stay overweight. MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 14 basis points in May. The compensation for prepayment risk (option cost) rose 2 bps on the month, but this was entirely offset by a 2 bps tightening in the option-adjusted spread (OAS). The most important issue for mortgage investors at the moment is when and how the Fed will cease the reinvestment of its MBS portfolio. We have written extensively on this topic in recent weeks,1 and through Fed communications have learned the following: The unwinding of the balance sheet will start before the end of this year (assuming the economic outlook does not deteriorate substantially) Both MBS and Treasury securities will be impacted The process will be "tapered" with monthly caps set on the amount of securities that will be allowed to run off. The caps will gradually increase according to a pre-set schedule. MBS OAS are already starting to look attractive, especially relative to Aaa-rated credit (Chart 4). But we are hesitant to move back into MBS at current levels. OAS have further upside relative to trends in net issuance (panel 4), and the increased supply from the end of Fed reinvestment will only add to the widening pressure. Remain underweight. Government-Related: Underweight Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 11 basis points in May, bringing year-to-date excess returns up to 86 bps. The Foreign Agency and Local Authority sectors outperformed by 18 bps and 38 bps, respectively. Meanwhile, the low-beta Domestic Agency and Supranational sectors outperformed by 7 bps and 9 bps, respectively. The Sovereign sector underperformed the Treasury benchmark by 12 bps on the month. Sovereigns underperformed in May even though the broad trade-weighted dollar depreciated by 1.4%. Similarly, Mexican debt - which carries the largest weighting in the Sovereign index - underperformed duration-equivalent Treasuries by 22 bps, even though the peso continued to appreciate versus the dollar (Chart 5). With U.S. growth likely to rebound following a weak Q1, the trade-weighted U.S. dollar should appreciate in the second half of this year. Meanwhile, our Emerging Markets Strategy thinks that Mexico's central bank could deliver another 25 bps rate hike, but it won't be long before tighter policy becomes a drag on consumer spending.2 The peso could stay well-bid for now, but the longer run trend is for a weaker peso versus the U.S. dollar. The Foreign Agency and Local Authority sectors continue to offer attractive spreads, after adjusting for credit rating and duration, compared to most U.S. corporate sectors. We continue to recommend overweight positions in Foreign Agencies and Local Authorities within an overall underweight allocation to the Government-Related Index. Municipal Bonds: Cut To Underweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 85 basis points in May (before adjusting for the tax advantage). The average Municipal / Treasury (M/T) yield ratio declined 8% on the month, and is now more than one standard deviation below its post-crisis mean. In a recent report,3 we noted that the current weakness in state & local government revenue growth mostly reflected the fall-out from the mid-2014 commodity price slump. As such, we expect that revenue growth will rebound in the months ahead and that state & local government net borrowing will decline. However, this eventuality is now fully discounted in M/T yield ratios (Chart 6, panel 3). Further, M/T yield ratios benefited from a steep decline in issuance during the past few months (bottom panel), and the recent uptick in visible supply suggests that the tailwind from declining issuance is about to shift. Factor in the uncertainty surrounding tax reform and a potential infrastructure program, and it is difficult to make the case for much tighter yield ratios. We recommend investors reduce municipal bond exposure to underweight (2 out of 5). Investors should continue to capture the premium in long-maturity munis relative to short maturities (panel 2), and also favor the debt of commodity-dependent states where tax revenues should grow more quickly. In particular, Aaa-rated Texas General Obligation bonds offer a premium of 14 bps versus the overall Aaa muni curve at the 10-year maturity point. The average premium offered by other Aaa-rated states is -0.6 bps. Treasury Curve: Favor 5-Year Bullet Over 2/10 Barbell Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve shifted lower and flattened in May. The 2/10 slope flattened 8 basis points and the 5/30 slope flattened 3 bps. For the second consecutive month yields remained stable out to the 2-year maturity point, but declined further out the curve. As stated on the first page of this report, the recent flattening of the Treasury curve indicates that the market expects the Fed will maintain a policy that is too restrictive for inflation to return to target. We think this is flat out wrong. Either core inflation will turn higher in the second half of this year, allowing the Fed to lift rates twice more in 2017. Or, core inflation will remain depressed. In the latter scenario, the Fed would adopt a more dovish policy stance until inflation starts to rise. In either case, the cost of inflation compensation at the long-end of the curve is not high enough, and it will cause the curve to steepen as it rises (Chart 7). We previously documented that the positive correlation between TIPS breakeven rates and the slope of the yield curve still holds during Fed rate hike cycles.4 We continue to recommend positioning for a steeper 2/10 curve by favoring the 5-year bullet versus a duration-matched 2/10 barbell. This trade returned 0 bps in May, but is still 26 bps in the money since inception on December 20, 2016. While this trade no longer benefits from the extreme cheapness of the 5-year bullet relative to the rest of the curve (panel 3), it will continue to outperform as TIPS breakevens widen and the curve steepens in the second half of the year. TIPS: Overweight Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS underperformed the duration-equivalent nominal Treasury index by 107 basis points in May. The 10-year TIPS breakeven rate fell 11 bps on the month and, at 1.79%, it remains well below its pre-crisis trading range of 2.4% to 2.5%. A series of disappointing inflation reports have led to weakness in TIPS breakevens so far this year. Year-over-year trimmed mean PCE inflation fell to 1.75% in April, all the way from a peak of 1.91% as recently as January (Chart 8). As we discussed in two recent reports,5 a Phillips Curve model- based on lagged inflation, the employment gap, non-oil import prices and inflation expectations - forcefully predicts that core inflation will trend higher for the remainder of the year (panel 4). In a base case scenario in which both the unemployment rate and the trade-weighted dollar remain flat at current levels, the model projects that core PCE inflation will exceed 2% by the end of this year. In fact, we find it difficult to create a set of reasonable economic assumptions that don't result in core PCE inflation at (or above) the Fed's 1.9% forecast by year end. While we anticipate a rebound in core inflation between now and the end of the year, if that rebound does not seem to be materializing by the end of the summer, the Fed is likely to adopt a more dovish policy stance. Such a policy shift would lend support to TIPS breakeven wideners. ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 19 basis points in May, bringing year-to-date excess returns up to +52 bps. The index option-adjusted spread for Aaa-rated ABS tightened 7 bps on the month, and remains well below its average pre-crisis level. In a recent report, we highlighted that consumer balance sheets are in their best shape since prior to the start of the housing bubble.6 As such, consumer ABS should remain a relatively low risk investment. However, some signs of stress are beginning to emerge, particularly in the sub-prime auto space. According to the Federal Reserve's Senior Loan Officer Survey, credit card lending standards tightened in Q4 of last year, but have since reverted into net easing territory (Chart 9). In contrast, auto loan lending standards continue to tighten and net losses on auto loans appear to have bottomed for the cycle. At least so far, auto ABS are not discounting much deterioration in credit quality. After adjusting for volatility, Aaa-rated auto ABS do not offer much of a spread pick-up relative to Aaa-rated credit card ABS (panel 3) and the spread differential between non-Aaa auto ABS and Aaa auto ABS has fallen to one standard deviation below its post-crisis mean. We continue to recommend that investors favor Aaa-rated credit cards over Aaa-rated auto loans within an overall overweight allocation to consumer ABS. Non-Agency CMBS: Underweight Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 33 basis points in May, bringing year-to-date excess returns up to +52 bps. The index option-adjusted spread for non-agency Aaa-rated CMBS tightened 4 bps on the month, but remains below its average pre-crisis level (Chart 10). Apartment and office building prices are growing strongly, but retail sector property prices have been close to flat during the past year (bottom panel). Tighter lending standards and falling demand also suggest that credit stress is starting to mount in the commercial real estate sector. So far, this stress has manifested itself in rising retail and office delinquency rates, while multi-family delinquencies remain low (panel 5). Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 31 basis points in May, bringing year-to-date excess returns up to +50 bps. The index option-adjusted spread for Agency CMBS tightened 5 bps on the month, and currently sits at 49 bps. The option-adjusted spread on Agency CMBS still looks attractive compared to other high-quality spread product: Agency MBS = 36 bps, Aaa consumer ABS = 39 bps, Agency bonds = 17 bps and Supranationals = 19 bps. We continue to recommend an overweight position in Agency CMBS. Treasury Valuation Chart 11Treasury Fair Value Models Treasury Fair Value Models Treasury Fair Value Models The current reading from our 2-factor Treasury model (which is based on Global PMI and dollar sentiment) places fair value for the 10-year Treasury yield at 2.49% (Chart 11). Our 3-factor version of the model, which also includes the Global Economic Policy Uncertainty Index, places fair value at 2.41%. The lower fair value results from the large spike in the uncertainty index last November, which has only been partially unwound. The U.S. PMI has dipped lower in recent months, but remains firmly entrenched above the 50 boom/bust line. Meanwhile, the Eurozone PMI continues to surge ahead. China's PMI is the real source of concern. It has recently dipped below 50, and there is a risk that tighter monetary policy could lead to further contraction in the near term (bottom panel).7 For further details on our Treasury models please refer to the U.S. Bond Strategy Weekly Report, "The Message From Our Treasury Models", dated October 11, 2016, available at usbs.bcaresearch.com At the time of publication the 10-year Treasury yield was 2.15%. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com Alex Wang, Research Analyst alexw@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "The Fed Doctrine", dated May 30, 2017, U.S. Bond Strategy Weekly Report, "Two Challenges For U.S. Policymakers", dated May 23, 2017, U.S. Bond Strategy Weekly Report, "The Payback Period In Corporate Bonds", dated April 11, 2017 and U.S. Bond Strategy / Global Fixed Income Strategy Special Report, "The Way Forward For The Fed's Balance Sheet", dated February 28, 2017. All available at usbs.bcaresearch.com 2 Please see Emerging Markets Strategy Weekly Report, "A Time To Be Contrarian", dated April 5, 2017, available at ems.bcaresearch.com 3 Please see U.S. Bond Strategy Weekly Report, "Will The Fed Stick To Its Guns?", dated May 16, 2017, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "The Yield Curve On A Cyclical Horizon", dated March 21, 2017, available at usbs.bcaresearch.com 5 Please see U.S. Bond Strategy Weekly Report, "The Fed Doctrine", dated May 30, 2017 and U.S. Bond Strategy Weekly Report, "Two Challenges For U.S. Policymakers", dated May 23, 2017. Both available at usbs.bcaresearch.com 6 Please see U.S. Bond Strategy Weekly Report, "The Fed Doctrine", dated May 30, 2017, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy / Global Fixed Income Strategy Weekly Report, "Past Peak Pessimism", dated May 9, 2017, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation Total Return Comparison: 7-Year Bullet Versus 2-20 Barbell (6-Month Investment Horizon) Current Recommendation