Banks

Bank stocks have experienced a sentiment-driven surge since the U.S. election, supported by expectations for higher interest rates. However, lost in the exuberance has been a marked deceleration in credit creation. Total bank loan growth has dropped to nil over the last three months, led by the previously booming C&I category. That is a sign that while businesses are expecting an economic improvement, they are not yet positioning for one via increasing working capital requirements. Coupled with increased bank staffing levels, the growth in bank loans-to-employment, a decent productivity proxy, has also dropped to zero. Importantly, the yield curve widening has taken a breather, which may be a catalyst for some profit-taking, especially if upcoming bank earnings results disappoint on the credit growth front. We are underweight this index. The ticker symbols for the stocks in this index are: BLBG: S5BANKX - WFC, JPM, BAC, C, USB, PNC, BBT, STI, MTB, FITB, CFG, RF, KEY, HBAN, CMA, ZION, PBCT. Banks: Will Higher Interest Rates Trump Weakening Loan Growth Banks: Will Higher Interest Rates Trump Weakening Loan Growth

Highlights The global 6-month credit impulse is now in its longest upcycle in a decade. Given also that bond yields have had their sharpest spike in a decade, we would not bet on the upcycle lasting much longer. Lean against the rise in bond yields and bank equities. Underweight the Eurostoxx600 versus the S&P500. Underweight the IBEX versus the Eurostoxx600. Feature A few days into the New Year, two over-arching economic questions are exercising our minds. Is the relationship between sharply higher bond yields and weaker bank credit creation still valid? And is the relationship between weaker bank credit creation and decelerating economic growth still valid? Chart of the WeekCredit Impulses Heading In Different Directions Credit Impulses Heading In Different Directions Credit Impulses Heading In Different Directions We suspect the answers are yes and yes. European Investors Must Think Globally, Not Locally Europe is not an investment island. Major European stock market indexes comprise large multinational companies whose sales and profits come from across the world. The upshot is that European stock markets almost always move up and down in tandem with other major world stock markets, such as the S&P500 and Nikkei225 (Chart I-2). Mainstream bond markets might seem to be more parochial, given that they are supposedly under the influence of the local central bank. But investors buy and sell high quality bonds as a global asset class. The upshot is that European bond markets also almost always move up and down in tandem with other major developed bond markets (Chart I-3). Chart I-2Major Equity Markets Move Together Major Equity Markets Move Together Major Equity Markets Move Together Chart I-3Major Bond Markets Move Together Major Bond Markets Move Together Major Bond Markets Move Together Hence, European investors must look first and foremost at global drivers. For us, the most important such driver is the global 6-month credit impulse - which sums the 6-month (dollar) credit impulses in the euro area, the United States and China. Does the global 6-month credit impulse have any predictive power? Yes. Chart I-4 shows that it has consistently led the 6-month cycle in the global government bond yield, which is a good proxy for the global growth cycle. Admittedly, this powerful predictive relationship weakened somewhat through 2013-14 during the most aggressive and distortive phase of worldwide QE. However, in the past couple of years, as QE has waned, the global 6-month credit impulse's predictive power has strongly re-asserted itself (Chart I-5). Chart I-4The Credit Impulse Leads ##br##The Global Growth 'Mini-Cycle' The Credit Impulse Leads The Global Growth 'Mini-Cycle' The Credit Impulse Leads The Global Growth 'Mini-Cycle' Chart I-5The Credit Impulse's Predictive ##br##Power Has Re-Asserted Itself The Credit Impulse's Predictive Power Has Re-asserted Itself The Credit Impulse's Predictive Power Has Re-asserted Itself In effect, the charts illustrate that whatever the structural economic backdrop, the global economy experiences a perpetual 'mini-cycle' lasting about 15-24 months. Higher bond yields (or credit restrictions) weaken the credit impulse; the weaker impulse then depresses growth; the depressed growth lowers bond yields; lower bond yields (or credit easing) strengthen the credit impulse; the stronger impulse then boosts growth; the boosted growth lifts bond yields; and back to the beginning... Remember, the credit impulse measures the growth in the credit flow. The important point to grasp is that the impulse can weaken even if the credit flow numbers themselves seem strong. For example, if the credit flow increased from $100bn to $150bn to $190bn it might appear to be growing very healthily. But actually, the impulse would have weakened from $50bn to $40bn, creating a headwind. Where are we in the perpetual mini-cycle? Today, the euro area credit impulse is losing momentum, while the U.S. impulse is waning. Which leaves China's rising credit impulse as the only component supporting the global credit impulse (Chart of the Week). But for how much longer? To repeat, it would just take the global credit flow to decelerate for the impulse to roll over. Now consider that high-quality bond yields have had their sharpest 6-month spike in a decade. And that the current 10 month upcycle in the global credit impulse already makes it the longest in a decade. Hence, we would not bet on this mini-upcycle lasting much longer. A Few Words On Our Credit Cycle Framework Our credit cycle framework has several features which uniquely define it. First, the framework focusses on bank credit. This is because the magic of fractional reserve banking allows a bank to create money and new spending power out of thin air. When somebody borrows from a bank, his bank account and spending power goes up, but nobody's spending power has to go down. In contrast, when somebody borrows by issuing a bond, it merely reallocates spending power from one person to another person. The borrower sees his bank account and spending power go up, but the lender sees his bank account and spending power go symmetrically down. Spending will rise to the extent that the borrower has a higher propensity to spend than the lender, but this may or may not be the case. Second, as already discussed, the framework focusses on the bank credit impulse - which measures the growth in the bank credit flow. This is just to compare apples with apples. Remember that GDP is itself a flow statistic. So the growth in GDP receives a contribution from the growth in the credit flow1 (and not from the flow itself). Third, the framework focusses on the 6-month bank credit impulse. We choose this periodicity because 6 months is about the time that it takes to for credit to be fully spent, thereby yielding the greatest predictive power from the credit impulse to economic activity. Fourth, the framework calculates the credit cycle using bank credit to the non-financial sector2 rather than the more commonly-quoted money supply statistics such as euro area M3. The simple reason is that not all loans generate economic activity. Bank to bank lending may stay trapped in the financial system. The money supply - which is on the liabilities side of the banks' balance sheet - would not pick up this distinction. As M3 captures all bank deposits, it would still be expanding rapidly, giving the false signal that demand should be growing. Hence, it is better to focus on bank lending - which is on the assets side of the banks' balance sheet - and only count lending that is likely to generate economic activity (Chart I-6). This reasoning only works if the official data on bank loans is accurate and complete. In China, this is unlikely to be the case, given its large shadow banking system. But unofficial shadow lending must eventually show up in the money supply. Therefore, exceptionally for the China sub-component, our credit cycle framework does prefer to use the money supply (Chart I-7). Chart I-6Our Euro Area Credit Impulse##br## Uses Bank Lending... Our Euro Area Credit Impulse Uses Bank Lending... Our Euro Area Credit Impulse Uses Bank Lending... Chart I-7...But Our China Credit Impulse ##br##Uses Money Supply ...But Our China Credit Impulse Uses Money Supply ...But Our China Credit Impulse Uses Money Supply A Few Words On Our Reductionist Framework We are also strong believers in Investment Reductionism. This philosophy stems from two guiding principles: Occam's Razor - which says that when there are competing explanations for the same effect, the simplest explanation is usually the best; and the Pareto Principle - which says that 80% of effects come from just 20% of causes.3 The important point is that most of the moves in most financial markets result from a very small number of over-arching macro drivers. To reiterate, Europe is not an investment island. Investment Reductionism means that much of asset allocation, sector selection, and regional and country allocation distills down to getting the global growth cycle right. The remaining charts should leave readers in no doubt. Chart I-8 shows that the global 6-month credit impulse leads the cyclical direction of the global bond yield, and thereby determines asset class selection. Chart I-9 then shows that the direction of bond yields determines sector selection: for example Banks versus Technology. Chart I-8Investment Reductionism Step 1: The Global##br## Credit Impulse Leads The Bond Yield Cycle Investment Reductionism Step 1: The Global Credit Impulse Leads The Bond Yield Cycle Investment Reductionism Step 1: The Global Credit Impulse Leads The Bond Yield Cycle Chart I-9Step 2: The Bond Yield ##br##Drives Sector Selection Step 2: The Bond Yield Drives Sector Selection Step 2: The Bond Yield Drives Sector Selection Chart I-10 and Chart I-11 then show that the sector selection of Banks versus Technology determines both the regional allocation of Eurostoxx600 versus S&P500, and the country allocation of IBEX versus Eurostoxx600. Chart I-10Step 3: Sector Selection Drives##br## Regional Allocation Step 3: Sector Selection Drives Regional Allocation Step 3: Sector Selection Drives Regional Allocation Chart I-11Step 4: Sector Selection Drives ##br##Country Allocation Step 4: Sector Selection Drives Country Allocation Step 4: Sector Selection Drives Country Allocation To sum up, the global 6-month credit impulse is now in its longest up-cycle in a decade, and bond yields have had their sharpest spike in a decade. Hence, we would not chase cyclicality at this juncture. Which means that on a 6-month horizon: Lean against the rise in bond yields and bank equities. Underweight the Eurostoxx600 versus the S&P500.4 Underweight the IBEX versus the Eurostoxx600. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 Equivalently, the credit impulse is the growth in the growth (second derivative) of the credit stock. 2 The non-financial sector includes households, (non-financial) firms and government. 3 Often known as the 80-20 rule. 4 BCA Strategists differ on this position. Fractal Trading Model* This week's trade is to express a tactical short in equities via Italy's MIB. An alternative market-neutral trade is to go short Italy's MIB and symmetrically long Hong Kong's Hang Seng. * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-12 Short Italy's MIB Short Italy's MIB Chart I-13 Short Italy's MIB / Long Hong Kong's Hang Seng Short Italy's MIB / Long Hong Kong's Hang Seng Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II_2Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Indicators To Watch - Bond Yields Interest Rate Chart II-5Indicators To Watch ##br##- Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch ##br##- Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-7Indicators To Watch##br## - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch##br## - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations Indicators To Watch - Interest Rate Expectations

Highlights The valuation discount on Italian banks still seems insufficient for the sector's excess NPLs. We expect a better long-term buying opportunity sometime next year. Stay underweight the MIB and IBEX versus the Eurostoxx600. Stay underweight the Eurostoxx600 versus the S&P500. Long Italian BTPs versus French OATs has quickly achieved its profit target. Now prefer long Spanish Bonos versus French OATs. Feature Assessing The Value In Italian Banks Investment reductionism says that the valuation of a bank distils down to three things: The expected size of the bank's assets. In the standard banking model, the dominant asset is the bank's loan book. The expected profitability of the bank's assets. In the standard banking model, the dominant driver of profitability is the net interest margin (the difference between the interest rate received on loans and the interest rate paid on deposits). The expected amount of equity capital required against the bank's assets. The equity capital must absorb the bank's loan losses but it also receives the profits. Increasing the amount of equity capital dilutes the profits over a larger number of shares, and thereby lowers the bank's share price. Chart of the WeekSpain And France have Raised €100bn Of Bank Equity Capital... Italy Has Not Spain And France have Raised €100bn Of Bank Equity Capital... Italy Has Not Spain And France have Raised €100bn Of Bank Equity Capital... Italy Has Not Today, the potential reward of owning Italian banks is that they trade at a large valuation discount. Admittedly, growth in assets and profit margins is likely to be anaemic. But Italian banks trade on a price to forward earnings multiple of less than 10. Not only does this seem cheap in absolute terms, it is a 25% discount to other European banks (Chart I-2). Chart I-2Italian Banks Trade At A Significant Discount bca.eis_wr_2016_12_15_s1_c2 bca.eis_wr_2016_12_15_s1_c2 Chart I-3Italian Bank NPLs Have Increased Sharply bca.eis_wr_2016_12_15_s1_c3 bca.eis_wr_2016_12_15_s1_c3 But the risk of owning Italian banks is that they still carry €170bn of un-provisioned non-performing loans (NPLs), which is likely to require a large - and dilutive - increase in equity capital. NPLs have increased much more sharply in Italy than in Spain or France (Chart I-3). But the more significant difference is that Italian banks have not yet raised equity capital as a cushion against their rising NPLs. Since 2009, Spanish banks and French banks have both increased their equity capital by more than €100bn. Over this same period, Italian banks have actually shrunk their equity capital (Chart of the Week). Given that Italian bank equity capital stands at €150bn, today's 25% valuation discount is pricing a dilutive increase in equity capital of around €50bn. Will this be a sufficient cushion? Chart I-4How Much Equity Capital Do Italian Banks Require? How Much Equity Capital Do Italian Banks Require? How Much Equity Capital Do Italian Banks Require? Our assessment is that it still might be insufficient. Our prudent benchmark is that the Italian banking sector lifts its equity capital to NPL multiple to the lowest coverage that the Spanish banking sector reached in recent years (Chart I-4). That would require Italy to emulate Spain and France and raise closer to €100bn of fresh bank equity capital. Also beware that if an undercapitalized bank cannot raise sufficient equity capital privately in the markets, there is a danger that its investors could suffer heavy losses. This is because the EU rules on state aid for banks changed at the start of 2016. The EU Bank Recovery and Resolution Directive (BRRD) allows a government to step in with a 'precautionary' capital injection only after a first-loss 'bail-in' of the bank's equity and bond holders. Hence, Italian banks are a potential buy if you believe €50bn of extra equity capital will fully alleviate concerns about the large stock of un-provisioned NPLs... and if you believe that the sector's plan to raise equity capital in the market will avoid any major mishap. Given global banks' strong recent bounce, we expect a better long-term buying opportunity sometime next year. Value Doesn't Help Pick Equity Markets Chart I-5Italy's MIB Looks Cheap bca.eis_wr_2016_12_15_s1_c5 bca.eis_wr_2016_12_15_s1_c5 The headline cheapness of Italian banks inevitably makes Italy's MIB look relatively cheap too (Chart I-5), especially given the Italian stock market's overweighting to banks. Some people suggest sector-adjusting stock market valuations to neutralize the dominating sector skews, thereby creating a truer picture of relative valuation. Adjusted for these sector skews, is a stock market cheap or expensive? This question may be of interest to academics, but it has very little practical relevance for stock market selection. Compared to France's CAC, Italy's MIB and Spain's IBEX are indeed cheaper mainly because of their large overweight to banks. But this cannot change the inescapable fact that this defining large overweight to banks is precisely what drives MIB and IBEX relative performance. Likewise, compared to the S&P500, the Eurostoxx600 is much cheaper mainly because of its overweight to banks combined with its large underweight to technology. But this cannot change the inescapable fact that this defining overweight to banks combined with large underweight to technology is precisely what drives Eurostoxx600 versus S&P500 relative performance. For the sceptics, the charts on page 5 should leave no doubt that everything else is largely irrelevant. The recent outperformance of banks is just a manifestation of the Trump reflation trade, nothing more, nothing less (Chart I-6). Indeed, most of the moves in financial markets over the past month reduce to the same trade in one guise or another. This reflation trade has gone too far too fast, and we would now lean against it. An underweight to banks necessarily means underweighting the MIB and IBEX (Charts I-7 and I-8). Chart I-6The Trump Reflation Trade Has Lifted Banks bca.eis_wr_2016_12_15_s1_c6 bca.eis_wr_2016_12_15_s1_c6 Chart I-7Banks Drive The MIB Relative Performance bca.eis_wr_2016_12_15_s1_c7 bca.eis_wr_2016_12_15_s1_c7 Chart I-8Banks Drive The IBEX Relative Performance bca.eis_wr_2016_12_15_s1_c8 bca.eis_wr_2016_12_15_s1_c8 An underweight to banks and overweight to technology necessarily means underweighting the Eurostoxx600 versus the S&P500 (Charts I-9 and I-10). Chart I-9Banks Versus Technology... bca.eis_wr_2016_12_15_s1_c9 bca.eis_wr_2016_12_15_s1_c9 Chart I-10...Drive Eurostoxx600 Versus S&P500 ...Drive Eurostoxx600 Versus S&P500 ...Drive Eurostoxx600 Versus S&P500 Assessing The Value In French, Spanish And Italian Bonds Turning to bonds, the market has deemed that Italian BTPs and Spanish Bonos are more risky investments than French OATs. Therefore BTPs and Bonos require a yield premium over OATs. But what exactly is this yield premium for? In the unlikely event that a large euro area country like Italy or Spain defaulted on its sovereign euro-denominated debt, the monetary union as it stands would be unable to withstand the losses. The euro would likely break up, causing each country to redenominate its bonds into its own new currency, which would then rise or fall against the other new currencies. Today's yield premium on BTPs and Bonos over OATs is simply the expected value of the (annualised) loss that would be suffered in that eventuality. And this expected loss equals the product of two terms: the annual probability of euro break up and the expected depreciation of a new Italian lira (or new Spanish peseta) versus a new French franc after such a break up In turn, the expected depreciation of the lira or peseta versus the franc would broadly equal the respective economy's accumulated competitiveness shortfall versus France. Which leads to a powerful conclusion. Spain has rapidly eroded its competitiveness shortfall versus France, and is on course for full convergence within a couple of years (Chart I-11). If the second term of the above product becomes zero, so too must the product itself. Meaning the yield premium on Bonos over OATS must converge to zero - irrespective of whether the euro survives or not. Chart I-11Spainish Competitiveness Will Soon Reconverge With French Competitiveness Spainish Competitiveness Will Soon Reconverge With French Competitiveness Spainish Competitiveness Will Soon Reconverge With French Competitiveness Stay long Spanish Bonos versus French OATs. In the case of Italy, a substantial shortfall in competitiveness versus France (and now Spain) does exist, justifying a structural yield premium in BTPs. But recently, this premium widened further because of a larger first term in the above product - a perceived increase in the annual probability of euro break up after the no vote in Italy's referendum on constitutional reform, and Prime Minister Renzi's subsequent resignation. However, as we argued in Italy: Asking The Wrong Question,1 fears of the political repercussions of a no vote were overdone. As the market has come to realise this, the BTP yield premium has quickly retraced most of its recent widening. Our long Italian BTP versus French OATs bond pair trade has achieved its profit target in just 10 days, and we are now closing it (see section below). Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 Published on December 1, 2016 and available at eis.bcaresearch.com Fractal Trading Model* This week's recommended trade is to buy gold. Long Gold Long Gold Long Gold * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Long Italian Government Bonds / Short French Government Bonds Long Italian Government Bonds / Short French Government Bonds Long Italian Government Bonds / Short French Government Bonds The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields bca.eis_wr_2016_12_15_s2_c1 bca.eis_wr_2016_12_15_s2_c1 Chart II-2Indicators To Watch - Bond Yields bca.eis_wr_2016_12_15_s2_c2 bca.eis_wr_2016_12_15_s2_c2 Chart II-3Indicators To Watch - Bond Yields bca.eis_wr_2016_12_15_s2_c3 bca.eis_wr_2016_12_15_s2_c3 Chart II-4Indicators To Watch - Bond Yields bca.eis_wr_2016_12_15_s2_c4 bca.eis_wr_2016_12_15_s2_c4 Interest Rate Chart II-5Indicators To Watch ##br## - Interest Rate Expectations bca.eis_wr_2016_12_15_s2_c5 bca.eis_wr_2016_12_15_s2_c5 Chart II-6Indicators To Watch ##br## - Interest Rate Expectations bca.eis_wr_2016_12_15_s2_c6 bca.eis_wr_2016_12_15_s2_c6 Chart II-7Indicators To Watch ##br## - Interest Rate Expectations bca.eis_wr_2016_12_15_s2_c7 bca.eis_wr_2016_12_15_s2_c7 Chart II-8Indicators To Watch ##br## - Interest Rate Expectations bca.eis_wr_2016_12_15_s2_c8 bca.eis_wr_2016_12_15_s2_c8

The scope of the revaluation in banks stocks on hopes for reduced regulatory constraints and the modest yield curve steepening has surprised us, but it would be dangerous to equate share price strength with the perception that underlying activity has reaccelerated. Indeed, the chart shows that overall bank loan growth is steadily decelerating, led by a cooling in C&I credit creation to its lowest rate since the great recession. While credit growth is far from recessionary levels, it is diverging negatively from what bank stocks might suggest. Importantly, banks have been adding to cost structures in recent months, and our gauge of bank sector productivity (bank loans/bank employment) is receding on a growth rate basis. This suggests that there could be a setback if fourth quarter profits do not validate the share price move. Rather than chase the bank group, our preference to play strength in the financial sector is through the asset manager and consumer finance groups, see the next Insight. The ticker symbols for the stocks in this index are: BLBG: S5BANKX-JPM, WFC, BAC, C, USB, PNC, BBT, STI, MTB, FITB, KEY, CFG, RF, HBAN, CMA, ZION, PBCT. bca.uses_in_2016_12_14_001_c1 bca.uses_in_2016_12_14_001_c1

Highlights Duration: An easing of financial conditions is likely necessary for recent improvements in U.S. economic growth to continue. As such, the uptrend in Treasury yields will pause in the near-term before resuming early next year. Corporate Bonds: The macro back-drop is turning marginally more positive for corporate spreads. C&I lending standards are no longer tightening and bank stocks have rallied significantly. Corporate Bonds: Spreads are too tight at the moment, even for an improving economic environment. Remain neutral (3 out of 5) on investment grade and underweight (2 out of 5) on high-yield for now. We are actively looking to add exposure to corporate credit from more attractive levels. Feature There is no question that the U.S. economy is on a firm footing heading into the New Year. Third quarter real GDP growth came in at a robust 3.2%, and the Atlanta and New York Fed tracking models currently forecast fourth quarter growth of 2.6% and 2.7%, respectively. This represents a marked acceleration from the average growth rate of 1.1% witnessed during the first two quarters of 2016. Forward-looking survey data are also pointing in the right direction. The ISM non-manufacturing survey reached 57.2 in November, its highest level since October 2015, while the expectations component of the University of Michigan Consumer Sentiment survey reached 88.9 in December, its highest level since January 2015 (Chart 1). The question for bond investors is how much of this good news is already reflected in Treasury yields. Higher Treasury yields and a stronger dollar have already led to a material tightening in some broad indexes of financial conditions, enough to exert a meaningful drag on U.S. growth (Chart 2). In fact, according to the Fed's FRB/US model, the recent interest rate and dollar moves could be expected to shave 1% from GDP over the next eight quarters. Chart 1Economic Tailwinds Economic Tailwinds Economic Tailwinds Chart 2Financial Conditions Must Ease bca.usbs_wr_2016_12_13_c2 bca.usbs_wr_2016_12_13_c2 The natural conclusion is that while some upside in Treasury yields is justified by an improving economic outlook, the bond selloff has proceeded too quickly and must pause in the near-term to prevent financial conditions from exerting an excessive drag on growth. Sentiment and positioning indicators also confirm that the uptrend in yields appears stretched (Chart 2, bottom two panels). As such, last week we tactically shifted our recommended portfolio duration allocation from 'below benchmark' to 'at benchmark'.1 We expect Treasury yields will grind higher next year, reaching a range of 2.8% to 3% by the end of 2017, but the selloff will proceed more gradually, in line with the acceleration in economic growth. A More Uncertain World The premise that the bond selloff has proceeded too quickly is confirmed by our Global PMI models of the 10-year Treasury yield. We track two versions of our Global PMI model. One is a 2-factor model based only on the Global PMI index and a survey of bullish sentiment toward the U.S. dollar. The intuition behind this model is that improving global growth contributes to a higher fair value Treasury yield. However, for a given level of global growth, increasingly bullish dollar sentiment applies downward pressure to yields. This is because a stronger dollar represents a tightening of monetary conditions, so that all else equal, a stronger dollar means we should expect fewer Fed rate hikes. The current fair value reading from this 2-factor model is 2.26%, meaning that the 10-year Treasury yield at 2.49% appears somewhat cheap (Chart 3). The second version of our Global PMI model is a 3-factor model which adds the Global Economic Policy Uncertainty Index (EPUI) as a third independent variable. All else equal, an increase in uncertainty about the economic outlook should depress the term premium in long-dated Treasury yields. The data appear to back-up this assertion, as the EPUI is negatively correlated with the 10-year Treasury yield over time. With the addition of the EPUI, our 3-factor model explains 84% of the variation in the 10-year Treasury yield since 2010, compared to 80% from our 2-factor model. The EPUI spiked last month, and as such, this version of the model suggests that fair value for the 10-year Treasury yield is only 1.82% (Chart 4). Chart 32-Factor Global PMI Model bca.usbs_wr_2016_12_13_c3 bca.usbs_wr_2016_12_13_c3 Chart 43-Factor Global PMI Model bca.usbs_wr_2016_12_13_c4 bca.usbs_wr_2016_12_13_c4 There are probably good reasons to overlook last month's spike in policy uncertainty. For one, the EPUI, created by Baker, Bloom and Davis,2 is largely constructed from algorithms that scan newspaper articles for keywords. They do not attempt to distinguish between economic news with bond-bearish or bond-bullish implications. Second, we have found that large spikes in uncertainty that do not coincide with deterioration in economic growth tend to mean-revert fairly quickly. This past summer's Brexit vote being a prime example. As a counterpoint, however, the negative correlation between the EPUI and the 10-year Treasury yield is quite robust (Chart 5), and historically, incidents of spiking policy uncertainty and rising Treasury yields have been few and far between. Since 1991, there have been 42 instances when the monthly increase in the EPUI exceeded one standard deviation. In those 42 months, the 10-year Treasury yield increased only 36% of the time, with last month's 53 basis point rise being by far the largest on record. We tend to view the reading from the 2-factor model as the more reasonable assessment of fair value in the current environment. But the spike in policy uncertainty does underscore why we should view the recent bond selloff skeptically. The recent selloff has, to a large extent, been predicated upon promises of fiscal stimulus that have yet to be delivered, from a President-elect who has shown himself to be highly unpredictable. In this environment, near-term caution is clearly warranted. Of course, this week the market's focus will at least temporarily turn away from fiscal policy and toward the Fed. We expect that the Fed will announce a 25 basis point increase in the fed funds rate tomorrow, but also that participants' interest rate projections will not change meaningfully. The FOMC will likely be much slower to react to promises of fiscal stimulus than the market. With the Fed's projected near-term path for interest rates already mostly discounted by the market (Chart 6), we could see a "dovish hike" from the Fed tomorrow coinciding with the near-term top in Treasury yields. Chart 5Economic Policy Uncertainty & Treasury Yields bca.usbs_wr_2016_12_13_c5 bca.usbs_wr_2016_12_13_c5 Chart 6A "Dovish Hike" Is In The Price bca.usbs_wr_2016_12_13_c6 bca.usbs_wr_2016_12_13_c6 Bottom Line: An easing of financial conditions is likely necessary for recent improvements in U.S. economic growth to continue. As such, the uptrend in Treasury yields will pause in the near-term before resuming early next year. A More Favorable Environment For Credit We frequently point to three main indicators that we use to assess the current stage of the credit cycle: Our Corporate Health Monitor (CHM) Monetary conditions relative to equilibrium C&I bank lending standards In a report3 published earlier this year we found that the performance of bank stocks relative to the overall market is another useful indicator (Chart 7). While the credit cycle is still very much in its late stages, recently, our indicators have been sending marginally more positive signals. The CHM remains deep in 'deteriorating health' territory and non-financial corporate balance sheets continue to lever-up aggressively. However, the indicator did inch slightly closer to 'improving health' territory in the third quarter due to an improvement in all six of its components (Chart 8). Make no mistake, trends in corporate balance sheet leverage are not supportive for corporate spreads. In fact, as we will explore in a future report, the recent divergence between rising leverage and tightening spreads is nearly unprecedented during the past 40 years. But at the margin, recent trends are less worrisome. Chart 7Credit Cycle Indicators Credit Cycle Indicators Credit Cycle Indicators Chart 8Corporate Health Monitor Components Corporate Health Monitor Components Corporate Health Monitor Components Box1: Corporate Health Monitor Components The BCA Corporate Health Monitor is a normalized composite of six financial ratios, calculated for the non-financial corporate sector as a whole. These six ratios are defined as follows: Profit Margins: After-tax cash flow as a percent of corporate sales Return on Capital: After-tax earnings plus interest expense, as a percent of capital stock Debt Coverage: After-tax cash flow less capital expenditures, as a percent of all interest bearing debt Interest Coverage: EBITDA (Earnings before interest, taxes, depreciation & amortization) divided by the sum of interest expense and dividends Leverage: Total debt as a percent of market value of equity Liquidity: Working Capital, excluding inventories, as a percent of market value of assets Second, although monetary conditions appear very close to our estimate of equilibrium, the recent steepening of the yield curve suggests that the market is revising its estimate of monetary equilibrium higher, leading to a de-facto easing of monetary conditions. In the long-run, with the Fed in the midst of a hiking cycle, this sort of easing is unlikely to persist. But, as we argued in a recent report,4 the bear steepening curve environment could continue in the first half of next year as the Fed is slow to respond to an improving economy. Third, C&I bank lending standards have fallen back to unchanged after having tightened for four consecutive quarters. This likely reflects less stress in the energy sector now that oil prices have rebounded. Fourth, bank stocks have rallied strongly alongside the steepening yield curve. To the extent that higher bank stock prices reflect lower future commercial loan delinquencies, then this trend should be viewed positively from the perspective of credit investors. To test the idea that bank stock performance might help us trade the corporate bond market, we take a look at the past six credit cycles, going back to 1975 (Chart 9). The bottom panel of Chart 9 shows the percent drawdown in relative bank equity performance from its peak during the most recent credit cycle. We define credit cycles as the periods between when the CHM crosses into 'improving health' territory. For example, we define the most recent credit cycle as beginning when the CHM fell into 'improving health' territory in 2002 and ending when it fell into 'improving health' territory in 2009. Shaded regions in Chart 9 show periods when the CHM is in 'deteriorating health' territory. Chart 9Bank Equity Drawdown & Corporate Bond Performance bca.usbs_wr_2016_12_13_c9 bca.usbs_wr_2016_12_13_c9 If we construct a trading strategy using the CHM alone, we can get fairly good results. We find that investment grade corporate bonds underperform duration-equivalent Treasury securities in 3 out of 6 instances, over a 12-month investment horizon, following the time when the CHM first crosses into deteriorating health territory, for an average excess return of -1.2% (Table 1). Table 1Corporate Bond Trading Rules: 12-Month Investment Horizon A Positive Signal From Bank Stocks A Positive Signal From Bank Stocks However, we find that this result can be improved if we also incorporate bank stock price performance. If we were to only reduce corporate bond exposure when the CHM was in deteriorating health territory and after the drawdown in bank equities exceeded 20%, then the position is still profitable in 3 out of 6 instances, but for a more negative average return of -1.9%. Further, if we were to wait for the drawdown in bank equities to surpass 30%, then the hit rate on our position improves to 3 out of 5 and the average return falls to -4.6%. We find similar results if we use a 6-month investment horizon (Table 2). In the current cycle, the drawdown in bank stocks breached 25% in February but has since reversed course, and it has not yet reached the 30% threshold. Our analysis suggests that corporate bond underperformance tends to persist for some time even after the drawdown in bank stocks exceeds 30%. Table 2Corporate Bond Trading Rules: 6-Month Investment Horizon A Positive Signal From Bank Stocks A Positive Signal From Bank Stocks Chart 10Corporate Spreads Are Too Low Corporate Spreads Are Too Low Corporate Spreads Are Too Low Bottom Line: The macro back-drop is turning marginally more positive for corporate spreads. We remain neutral (3 out of 5) on investment grade and underweight (2 out of 5) on high-yield for now, due to poor starting valuation (Chart 10). But we are looking for an opportunity to upgrade from more attractive spread levels in the next couple of months. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Portfolio Allocation Summary, "Too Far Too Fast, But The Bond Bear Is Still Intact", dated December 6, 2016, available at usbs.bcaresearch.com 2 For further details on the construction of this index please see www.policyuncertainty.com 3 Please see U.S. Bond Strategy Weekly Report, "Lighten Up On Duration", dated February 16, 2016, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, "Toward A Cyclical Sweet Spot?", dated November 22, 2016, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification

Highlights ECB QE has pushed the euro area's Target2 banking imbalance to an all-time high. Thereby, QE has raised the cost of euro break-up. The ECB must dial down QE because the Target2 banking imbalance is directly related to the size of asset purchases. Core euro area sovereign bonds offer poor relative value in the government bond universe. Long Italian BTPs / short French OATs is now appropriate as a tactical position. Italian bank investors might have to suffer more pain before Brussels ultimately allows a public rescue. Feature "We've eliminated fragmentation in the euro area." Mario Draghi, speaking on October 20, 2016 Mario Draghi is wrong. QE was meant to reduce economic and financial fragmentation within the euro area. But in one important regard, it has done the exact opposite. In an un-fragmented monetary union, banking system liquidity would be spread evenly across the euro area. Unfortunately, the trillions of euros of QE liquidity created by the ECB has concentrated in four northern European countries: Germany, the Netherlands, Luxembourg and Finland (but interestingly, not France). This extreme fragmentation is captured in the euro area's Target2 banking imbalance (Box I-1), which is now at an all-time high (Chart of the Week). Box 1: What Is Target2? Target2 stands for Trans-European Automated Real-time Gross settlement Express Transfer system. It is the settlement system for euro payment flows between banks in the euro area. These payment flows result from trade or financial transactions such as deposit transfers, sales of financial assets or debt repayments. If the banking system in one member country has more payment inflows than outflows, its national central bank (NCB) accrues a Target2 asset vis-à-vis the ECB. Conversely, if the banking system has more outflows than inflows, the respective NCB accrues a Target2 liability. Target2 balances therefore show the cumulative net payment flows within the euro area. Chart of the WeekQE Has Pushed The Euro Area's Target2 Imbalance To An All-Time High ECB QE Raises The Cost Of Euro Break-Up ECB QE Raises The Cost Of Euro Break-Up To be absolutely clear, this geographical polarization of bank liquidity is not deposit flight in the strictest sense (Chart I-2). Investors are simply using the ECB's €80bn of monthly bond purchases to offload their Italian, Spanish and Portuguese bonds to the central bank, and hold the received cash in banks in perceived haven countries. Nevertheless, ECB QE has unwittingly facilitated a geographical polarization of bank liquidity more extreme than in the darkest days of 2012 (Chart I-3). Chart I-2No Funding Stresses At The Moment bca.eis_wr_2016_12_08_s1_c2 bca.eis_wr_2016_12_08_s1_c2 Chart I-3Target2 Imbalances Are The Result Of QE Target2 Imbalances Are The Result Of QE Target2 Imbalances Are The Result Of QE QE Has Exposed Euro Area Banking Fragmentation To understand how this polarization has arisen, it is necessary to grasp how Eurosystem accounting works. The following section is necessarily technical, but stick with it because it is important. The ECB delegates its QE sovereign bond purchases to the respective national central bank (NCB): the Bundesbank buys German bunds, the Bank of France buys OATs, the Bank of Italy buys BTPs, and so on. When the Bank of Italy buys a BTP from, say, an Italian investor, the investor gives up the bond, but simultaneously receives a corresponding asset - cash. If the investor then deposits this cash at an Italian bank, say Unicredit, then Unicredit would have a new liability - the investor deposit. But in line with Eurosystem accounting, Unicredit would simultaneously receive a corresponding credit at its NCB, the Bank of Italy.1 Completing the accounting circle, the Bank of Italy would now have a new liability - the Unicredit claim, but it would also have a corresponding asset - the BTP that it has just bought. Therefore, all three accounts would be in perfect balance (see Figure I-1). Figure I-1Italian Investor Sells A BTP To The Bank Of Italy And Deposits The Cash At Unicredit ECB QE Raises The Cost Of Euro Break-Up ECB QE Raises The Cost Of Euro Break-Up Now consider what happens if the Italian investor deposits the cash not at Unicredit, but at a German bank, say Commerzbank. In this case, it would be the Bundesbank that had a new liability - the Commerzbank claim. However, the Bundesbank would not have a corresponding asset. Conversely, the Bank of Italy would have a new asset - the BTP, but without a corresponding liability. In order to balance these Eurosystem accounts, the Bundesbank would accrue a Target2 asset vis-à-vis the ECB, while the Bank of Italy would accrue an equal and opposite Target2 liability (see Figure I-2). Figure I-2Italian Investor Sells A BTP To The Bank Of Italy And Deposits The Cash At Commerzbank ECB QE Raises The Cost Of Euro Break-Up ECB QE Raises The Cost Of Euro Break-Up Essentially, the Target2 imbalance captures the mismatch between a Bundesbank liability denominated in 'German' euros and a corresponding Bank of Italy asset denominated in 'Italian' euros. Aggregated over the whole euro area, these imbalances now amount to more than €1 trillion. Does any of this Eurosystem accounting gymnastics really matter? No, as long as the monetary union holds together and the 'German' euro equals the 'Italian' euro. But if Germany and Italy started using different currencies, then suddenly the Target2 imbalances would matter enormously. This is because the Bundesbank liability to Commerzbank would be redenominated into Deutschemarks, while the Bank of Italy asset would be redenominated into lira. Hence, the ECB might end up with much larger liabilities than assets. In which case, any shortfall would have to be borne by the ECB's shareholders - essentially, euro area member states pro-rata to GDP. The ECB Must Dial Down QE Unlike in the depths of the euro debt crisis, the current Target2 imbalances do not reflect deposit flight. Rather, they are the direct result of ECB QE. Nonetheless, the indisputable fact is that QE has increased the cost of euro break-up. And another six or more months of QE will just add to this cost. Some people might argue that by increasing the cost of a divorce, an actual split becomes less likely. But this reasoning is weak. As we have seen in this year's polling victories for Brexit and President-elect Trump, the biggest risk comes from a populist backlash against the status quo. And populist backlashes do not stop to do detailed cost benefit analysis. Although the ECB is unlikely to broadcast the unintended side-effects of its policy, it must be acutely aware that the costs of QE are rising while its benefits are diminishing. Given that the Target2 imbalances are directly related to the size of asset purchases, the ECB needs to indicate its intention to dial down QE purchases. And if it does need to loosen policy again in the future, it might be better off emulating the Bank of Japan - in targeting a yield rather than an asset purchase amount. The 6-9 month investment implication is that core euro area sovereign bonds offer poor relative value in the government bond universe. And within the core euro area, perhaps French OATs offer the least relative value. OATs are priced as haven sovereign bonds, yet interestingly Target2 imbalances suggest that banking liquidity flows do not regard France as a haven in the same way as Germany (Chart I-4 and Chart I-5). Chart I-4French OATs Are Priced ##br##As Haven Bonds... bca.eis_wr_2016_12_08_s1_c4 bca.eis_wr_2016_12_08_s1_c4 Chart I-5...But Banking Liquidity Flows Do Not ##br##Regard France As A Haven bca.eis_wr_2016_12_08_s1_c5 bca.eis_wr_2016_12_08_s1_c5 Another implication is that the euro should be stable or stronger against a basket of other developed economy currencies. Indeed, expect euro/pound to lurch up in the first half of next year when the U.K. government triggers Article 50 of the Lisbon Treaty to formally begin Brexit negotiations. Only then will the EU27 reveal its own negotiating strategy, and it is highly unlikely to be a sweet deal for the U.K. Italian Referendum Result: A Postscript The financial markets have shrugged off the Italian public's resounding "no" to constitutional reform, and rightly so. The current constitution, created in the aftermath of the Second World was designed to prevent a repeat of a populist like Benito Mussolini gaining power. Irrespective of whether the next General Election is in 2017 or 2018, the no vote actually reduces political tail-risk. A tactical position that is long Italian BTPs and short French OATs is now appropriate. As we discussed last week in Italy: Asking The Wrong Question the bigger issue is how Italy will unburden its banks of its non-performing loans (NPLs). Monte de Paschi's efforts at raising equity are baby steps in the right direction. But Monte de Paschi's €23 billion of sour loans2 are just the tip of Italy' NPL iceberg, which sizes up at €320 billion in gross terms and €170 billion net of provisions. These numbers, expressed as a share of GDP, show striking parallels with peak NPLs in Spain's banking system (Chart I-6 and Chart I-7). Spain ultimately unburdened its banks with a government bailout. Italy may have to do the same. But this will require Brussels to let Italy bend the EU's new bail-in rules for troubled and failing banks. Chart I-6Spain Unburdened Its Banks ##br##With A Government Bailout... bca.eis_wr_2016_12_08_s1_c6 bca.eis_wr_2016_12_08_s1_c6 Chart I-7...Italy May Ultimately##br## Do The Same bca.eis_wr_2016_12_08_s1_c7 bca.eis_wr_2016_12_08_s1_c7 The danger for investors is that Italian bank equity and bond holders might have to suffer more pain before Brussels relents. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 Unicredit and all other commercial banks use their accounts at their NCLs to make interbank payments. 2 MPS NPLs amount to €45bn in gross terms and €23bn net of provisions. Fractal Trading Model* Bucking the synchronized sell-off in global bonds, Greek sovereign bonds have actually rallied strongly in the last three months. But this rally could be near exhaustion, warranting a countertrend position. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-8 bca.eis_wr_2016_12_08_s1_c8 bca.eis_wr_2016_12_08_s1_c8 * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields bca.eis_wr_2016_12_08_s2_c1 bca.eis_wr_2016_12_08_s2_c1 Chart II-2Indicators To Watch - Bond Yields bca.eis_wr_2016_12_08_s2_c2 bca.eis_wr_2016_12_08_s2_c2 Chart II-3Indicators To Watch - Bond Yields bca.eis_wr_2016_12_08_s2_c3 bca.eis_wr_2016_12_08_s2_c3 Chart II-4Indicators To Watch - Bond Yields bca.eis_wr_2016_12_08_s2_c4 bca.eis_wr_2016_12_08_s2_c4 Interest Rate Chart II-5Indicators To Watch ##br##- Interest Rate Expectations bca.eis_wr_2016_12_08_s2_c5 bca.eis_wr_2016_12_08_s2_c5 Chart II-6Indicators To Watch##br## - Interest Rate Expectations bca.eis_wr_2016_12_08_s2_c6 bca.eis_wr_2016_12_08_s2_c6 Chart II-7Indicators To Watch ##br##- Interest Rate Expectations bca.eis_wr_2016_12_08_s2_c7 bca.eis_wr_2016_12_08_s2_c7 Chart II-8Indicators To Watch##br## - Interest Rate Expectations bca.eis_wr_2016_12_08_s2_c8 bca.eis_wr_2016_12_08_s2_c8

The financials sector has cheered President-elect Trump's anticipated policy changes as if they will all be enacted within very short order. While the yield curve has steepened modestly, the widening pales in comparison with the violence of the relative performance rise in financial share prices. We caution against extrapolating recent sector gains. If yields rise much further, they will be in danger of moving ahead of the rate of GDP growth, as the latter is unlikely to benefit from any fiscal stimulus for at least a few quarters. Financials (and the broad market), have trouble when this measure of liquidity tightens. The financial sector is already facing a cooling in the pace of credit creation and a decline in credit quality: financials sector ratings migration is steadily deteriorating. The bottom line is that the upward spike in financials sector relative performance is due for a breather, and only neutral weightings are recommended. Financials Overreaction Is Due For A Pause Financials Overreaction Is Due For A Pause

Highlights Eurostoxx600 outperformance versus the FTSE100 reduces to: will the euro weaken against the pound? Stay neutral in the Eurostoxx600 versus the FTSE100. Eurostoxx600 outperformance versus the Nikkei225 reduces to: will the euro weaken against the yen? Overweight the Eurostoxx600 versus the Nikkei 225. Eurostoxx600 outperformance versus the S&P500 reduces to: will European Banks outperform U.S. Technology? Underweight the Eurostoxx600 versus the S&P500. Feature 2016 is the year of the political shock. But for investors, some things have stayed faithfully the same. Chart of the WeekEurostoxx Vs. Nikkei Reduces To: Will Euro/Yen Weaken? Eurostoxx Vs. Nikkei Reduces To: Will Euro/Yen Weaken? Eurostoxx Vs. Nikkei Reduces To: Will Euro/Yen Weaken? Last week's report From Berlin Wall To Mexican Wall explained how the fall of the Berlin Wall in 1989 ushered in a great era of globalization - an era in which goods, services, capital and people have moved around the world more and more freely. Chart I-2The Globalization Of Stock Markets bca.eis_wr_2016_11_24_s1_c2 bca.eis_wr_2016_11_24_s1_c2 For investors, one major upshot is that the world's biggest companies have also become more and more globalized. The leading European stock market indexes - Eurostoxx600, FTSE100, DAX30, CAC40 and many other national indexes - are now just a collection of multinational companies with a global footprint. The same applies to major indexes outside Europe, such as the Nikkei225 and S&P500. Before the era of globalization, many companies had little exposure to economies outside their country or region of domicile. Unsurprisingly, in the 1980s, a German bank share price was more correlated with the rest of the German stock market than it was with a U.S. bank share price. But today, a large proportion of sales and profits are sourced globally. The German bank share price is now more correlated with the U.S. bank share price than it is with the rest of the German stock market! (Chart I-2) This begs the question: if Brexit and President-elect Trump are ushering in a great era of anti-globalization, will the major indexes become parochial once again? The answer is perhaps, but it will be a slow process - even assuming that the anti-globalization rhetoric does fully materialize. Sometimes, Stock Market Allocation Reduces To A Currency View For the time being, one obvious distinction between the major indexes will remain instrumental in driving performance differences. The Eurostoxx600 is denominated in euros, the FTSE100 in pounds, the Nikkei225 in yen, and the S&P500 in dollars. However, the constituent companies' sales and profits are denominated in a mixture of major global currencies, or in dollars. So all else being equal, if the local currency weakens - in other words, if other global currencies strengthen versus the local currency - then index profits will rise in local currency terms. Meaning the index value must go up. And if the local currency strengthens, the index value must go down. Simplistic as it sounds, some important asset allocation decisions just reduce to a bi-lateral currency view. Chart I-3 clearly shows that Eurostoxx600 versus FTSE100 relative performance reduces to a simple question: will the euro weaken against the pound? If so, the Eurostoxx600 will outperform the FTSE100. And vice-versa. Clearly, the outlook for euro/pound has been an important question this year, and will be an equally important question next year. Chart I-3Eurostoxx Vs. FTSE Reduces To: Will Euro/Pound Weaken? bca.eis_wr_2016_11_24_s1_c3 bca.eis_wr_2016_11_24_s1_c3 Likewise, the Chart of the Week clearly shows that Eurostoxx600 versus Nikkei225 relative performance reduces to a similar simple question: will the euro weaken against the yen? If so, the Eurostoxx600 will outperform the Nikkei225. And vice-versa. Sometimes, Stock Market Allocation Reduces To A Sector View But in the case of the Eurostoxx600 versus the S&P500, relative performance does not reduce to the direction of euro/dollar. Since mid-2014, the euro has weakened substantially versus the dollar, yet the Eurostoxx600 has underperformed the S&P500. This is because another factor drives this relative performance pair (Chart I-4 and Chart I-5). Chart I-4Eurostoxx Vs. S&P500 Does Not ##br##Depend On Euro/Dollar... bca.eis_wr_2016_11_24_s1_c4 bca.eis_wr_2016_11_24_s1_c4 Chart I-5...Eurostoxx Vs. S&P500 Does Depend ##br##On Banks Vs. Technology bca.eis_wr_2016_11_24_s1_c5 bca.eis_wr_2016_11_24_s1_c5 Although major indexes are a collection of multinational companies, it doesn't follow that the sector exposures of these indexes will be the same. Comparing the Eurostoxx600 with the S&P500, the Eurostoxx600 has a marked overexposure to Banks and an especially marked underexposure to Technology (Table I-1). Table I-1Eurostoxx Vs. S&P500 = Overweight##br## Banks, Underweight Technology More Investment Reductionism More Investment Reductionism Banks comprise 13% of the Eurostoxx600 market capitalization but only 6% of the S&P500. On the flipside, Technology comprises just 4% of the Eurostoxx600 market capitalization but a very substantial 21% of the S&P500. To repeat, multinational company share prices today are more correlated with their global sector than with their domestic stock market of listing. Recently, this has been true even for U.S. Banks - which amazingly have shown a higher correlation with European Banks than with the rest of the U.S. stock market. It follows that when two indexes are distinguished by large sector skews, these sector skews will drive relative performance. Our Special Reports Picking Countries The Right Way 1 Parts 1, 2 and 3 showed that this is the case for most head to head stock market comparisons within Europe. It is also the case for the Eurostoxx600 versus the S&P500. Put simply, for the Eurostoxx600 to outperform the S&P500 on a sustained basis, Banks must outperform Technology on a sustained basis. Or to be more precise, European Banks must outperform U.S. Technology (Chart I-6). Chart I-6Eurostoxx Vs. S&P500 Reduces To: Will European Banks Outperform U.S. Technology? Eurostoxx Vs. S&P500 Reduces To: Will European Banks Outperform U.S. Technology? Eurostoxx Vs. S&P500 Reduces To: Will European Banks Outperform U.S. Technology? Applying Reductionism To The Eurostoxx600 We can now apply investment reductionism to position the Eurostoxx600 against three other major indexes: the FTSE100, the Nikkei225 and the S&P500. 1. Eurostoxx600 outperformance versus the FTSE100 reduces to: will the euro weaken against the pound? For the foreseeable future, the euro/pound exchange rate hinges on the perceived severity of Brexit. In this regard, there is unlikely to be meaningful new information until the U.K. Supreme Court delivers its verdict on the legal process that the U.K. government must follow. The verdict is due in January. So for the time being, it is appropriate to stay neutral in the Eurostoxx600 versus the FTSE100. Eurostoxx600 outperformance versus the Nikkei225 reduces to: will the euro weaken against the yen? 2. The euro/yen exchange rate hinges on ECB/BoJ relative monetary policy. Given that the BoJ made its bold policy move a few months ago, the focus now is on whether the ECB will continue with QE beyond March 2017. Chart I-7European Banks Do Not Offer An Especially##br## Large Discount To U.S. Technology European Banks Do Not Offer An Especially Large Discount To U.S. Technology European Banks Do Not Offer An Especially Large Discount To U.S. Technology The minutes of the ECB's most recent policy meeting provide some clues. On the one hand, the central bank cautioned on the unintended consequences of extended QE: "The possible side effects of the low interest rate environment and the range of non-standard measures in place on the longer-term intermediation capacity of banks and other financial institutions had to be further examined" On the other hand, the ECB emphasised: "(QE) was set to run... in any case until the ECB saw a sustained adjustment in the path of inflation consistent with its inflation aim... underlying inflation, however, continued to lack clear signs of a convincing upward trend." On this basis, it seems that the ECB will extend its QE program beyond March 2017, as well as give a strong commitment to keep policy rates anchored. But the recent underperformance of the Eurostoxx600 versus Nikkei225 has discounted a sizable strengthening of euro/yen. It is appropriate to lean against this and overweight the Eurostoxx600 versus the Nikkei225. Eurostoxx600 outperformance versus the S&P500 reduces to: will European Banks outperform U.S. Technology? Again, the minutes of the ECB's most recent policy meeting perfectly summarized the environment for European banks: "Ongoing structural challenges to banks' balance sheets, notably arising from still high levels of non-performing loans (NPLs) in parts of the euro area banking sector, in conjunction with regulatory challenges (BRRD), and the weakness in profitability were seen to pose a risk to the transmission of monetary policy and a further recovery in credit dynamics" Or as we recently put it,2 European bank investors are fighting three long-term headwinds: BRRD, NPLs and NIRP. Yet on a price to forward earnings multiple, European Banks do not offer an especially large discount to U.S. Technology (Chart I-7). Therefore, investment reductionism says it is appropriate to underweight the Eurostoxx600 versus the S&P500. Dhaval Joshi, Senior Vice President European Investment Strategy dhaval@bcaresearch.com 1 Please see the three European Investment Strategy Special Reports 'Picking 5 European Countries The Right Way' November 13, 2014, 'Picking Countries The Right Way: Part 2' March 26, 2015 and 'Picking Countries The Right Way: Part 3' November 12, 2015, available at eis.bcaresearch.com 2 Please see the European Investment Strategy Weekly Report "All Roads Lead To Banks", dated October 6, 2016, available at eis.bcaresearch.com Fractal Trading Model* The recent sharp moves in markets offer another opportunity for a long plantinum / short palladium pair-trade. A similar opportunity on October 6 successfully signaled a 13% countertrend move. For any investment, excessive trend following and groupthink can reach a natural point of instability, at which point the established trend is highly likely to break down with or without an external catalyst. An early warning sign is the investment's fractal dimension approaching its natural lower bound. Encouragingly, this trigger has consistently identified countertrend moves of various magnitudes across all asset classes. Chart I-8 bca.eis_wr_2016_11_24_s1_c8 bca.eis_wr_2016_11_24_s1_c8 * For more details please see the European Investment Strategy Special Report "Fractals, Liquidity & A Trading Model," dated December 11, 2014, available at eis.bcaresearch.com The post-June 9, 2016 fractal trading model rules are: When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. Use the position size multiple to control risk. The position size will be smaller for more risky positions. Fractal Trading Model Recommendations Equities Bond & Interest Rates Currency & Other Positions Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields bca.eis_wr_2016_11_24_s2_c1 bca.eis_wr_2016_11_24_s2_c1 Chart II-2Indicators To Watch - Bond Yields bca.eis_wr_2016_11_24_s2_c2 bca.eis_wr_2016_11_24_s2_c2 Chart II-3Indicators To Watch - Bond Yields bca.eis_wr_2016_11_24_s2_c3 bca.eis_wr_2016_11_24_s2_c3 Chart II-4Indicators To Watch - Bond Yields bca.eis_wr_2016_11_24_s2_c4 bca.eis_wr_2016_11_24_s2_c4 Interest Rate Chart II-5Indicators To Watch ##br##- Interest Rate Expectations bca.eis_wr_2016_11_24_s2_c5 bca.eis_wr_2016_11_24_s2_c5 Chart II-6Indicators To Watch##br## - Interest Rate Expectations bca.eis_wr_2016_11_24_s2_c6 bca.eis_wr_2016_11_24_s2_c6 Chart II-7Indicators To Watch##br## - Interest Rate Expectations bca.eis_wr_2016_11_24_s2_c7 bca.eis_wr_2016_11_24_s2_c7 Chart II-8Indicators To Watch ##br##- Interest Rate Expectations bca.eis_wr_2016_11_24_s2_c8 bca.eis_wr_2016_11_24_s2_c8

Highlights A central bank cannot control/target the quantity and price of money simultaneously. For the past few years, China's central bank has silently moved away from controlling money growth toward targeting interest rates. As such, the reserve requirements imposed on banks have not and will not be a constraint on Chinese commercial banks' ability to lend and create money if the PBoC continues to supply banks with reserves "on demand." China's banks have created too many RMBs (broad money/deposits) and the PBoC has accommodated them. Such enormous supply of RMBs and mainland households' and companies' desire to get rid of their RMBs will lead to further yuan depreciation. Continue shorting the RMB and Asian currencies versus the U.S. dollar. Re-instate a short Colombian peso trade; this time against an equal-weighted basket of the U.S. dollar and the Russian ruble. Feature Following our October 26 Special Report titled, "Misconceptions About China's Credit Excesses",1 some clients have asked us how our analysis squares with fact that the People's Bank of China (PBoC) conducts its monetary policy using a reserve requirement ratio. The relevant question being, why would the PBoC's reserve requirements not limit commercial banks' ability to create money/credit? In that Special Report, we wrote: "A commercial bank is not constrained in loan origination by its reserves at the central bank if the latter supplies liquidity (reserves) to commercial banks "on demand." Given PBoC lending to banks has surged 5.5-fold over last three years (Chart I-1), we concluded that the reserve requirement ratio had, for all intents and purposes, lost its meaning in China. In this week's report we elaborate on this issue in detail. The main implication of our analysis today reinforces our conclusion from the previous report: namely, China's commercial banks have expanded credit enormously, and the PBoC has accommodated it. With respect to financial market implications, there are simply too many RMBs (broad money/deposits) in the system (Chart I-2). Chinese households and companies can instinctively sense this, and are opting to move their wealth into real assets, such as real estate, or foreign currencies. Hence, the oversupply of RMBs will continue to weigh on China's exchange rate, which will depreciate much further. We expect the US$/CNY to reach 7.8-8 over the next 12 months. Chart I-1The PBoC Has Provided Banks With Liquidity 'On Demand' bca.ems_sr_2016_11_23_s1_c1 bca.ems_sr_2016_11_23_s1_c1 Chart I-2There Are Too Many RMBs Floating Around bca.ems_sr_2016_11_23_s1_c2 bca.ems_sr_2016_11_23_s1_c2 Targeting Either The Quantity Or The Price Of Money Any central bank can target and control either the quantity of money or the price of money, but not both simultaneously. This holds true for any monopolist supplier of any good/service that does not have control over the demand curve. A demand curve for money is the function that ties the quantity demanded at various price points (the price being interest rates). Central banks - being monopolist suppliers of money, but unable to control money demand - must choose between controlling either the quantity of money or the price of money. The system of required reserves (RR) is a tool to control money supply (the quantity of money). When central banks reinforce the RR ratio, interbank interest rates typically swing enormously and often deviate considerably from the target policy rate (Chart 1). For example, when commercial banks expand loans too much and lack sufficient reserves at the central bank, they must borrow from the interbank market and thereby bid up interbank rates- i.e., short-term interest rates rise. This in turn restrains credit demand or the willingness to lend, and eventually reduces money growth. The opposite also holds true. When a central bank wants to target interest rates (the price of money), it cannot control money supply. To ensure that interbank/money market rates stay close to the policy rate - i.e., to reinforce its interest rate target - a central bank should provide the banking system with reserves "on demand." In other words, when interbank rates rise above the target policy rate, a central bank should inject sufficient liquidity into the system to bring interest rates down. Similarly, when interbank rates fall below the target policy rate, a central bank should withdraw enough liquidity from the banking system to assure interbank rates rise converging to its target policy rate. By supplying commercial banks with reserves (high powered money) "on demand" - i.e., providing as much reserves as they need - a central bank is de facto failing to enforce reserve requirements. As such, the central bank is giving up control over money creation. By and large, RRs lose their effectiveness if a central bank provides commercial banks with as much reserves as they request. In short, when a central bank opts for targeting interest rates, it cannot steer monetary aggregates - i.e., RRs and RR ratios lose their meaning. In the 1970s and 1980s, most central banks in advanced countries targeted money supply to achieve their policy goals such as inflation and sustainable economic growth. However, starting in the early 1990s, developed nations' central banks (the Federal Reserve, the Bank of England, the Bank of Canada, the Swiss National Bank and others) began to move away from controlling money supply (monetary aggregates) and toward targeting interest rates. Individual banks' limitations to borrow from the central bank often rests with the availability of collateral. So long as a commercial bank has eligible collateral (often government bonds), it can access central bank funding. This is true for Chinese commercial banks too. Bottom Line: Monetary authorities cannot control/target the quantity and price of money simultaneously. The Money Multiplier In An Interest Rate Targeting System When a central bank opts for targeting interest rates, commercial banks can originate an unlimited amount of loans and demand the central bank provide additional reserves, as long as they have eligible collateral. This corroborates our point from our previous report that a commercial bank's loan origination is not constrained by its reserves at the central bank if the latter supplies liquidity (reserves) "on demand." In a fractional reserve system, the ability of commercial banks to create loans/money is defined by a money multiplier. A potential ceiling for a money multiplier (MM) is calculated as: MM = (1 / RR ratio) For example, when the RR ratio is 10%: The money multiplier MM = (1 / 0.1) = 10 In effect, the banking system can create up to 10 times more money/loans/deposits per one dollar of reserves. Under the current system of interest rate targeting – which has prevailed among most developed countries since the early 1990s and more recently in China (more on China below) – we can think of the RR ratio as heading towards zero because central banks provide banks with almost unlimited liquidity (reserves). The RR ratio is not zero because there are still limitations on banks' ability to borrow from central banks due the availability (or lack thereof) of eligible collateral or compliance with Basel III requirements. Yet as the RR ratio gets smaller in size, its reciprocal (1 / RR ratio) becomes very large (not infinite, but a plausibly very large number). Overall, when a central bank targets interest rates, the ceiling of the money multiplier is not set by the central bank. Rather, the money multiplier is de facto determined by commercial banks' willingness to originate loans. Thus, the money multiplier can potentially be very high when animal spirits among bankers and borrowers run wild. Consequently, the points discussed in our Special Report titled, "Misconceptions About China's Credit Excesses"2 - namely that commercial banks create loans/money/deposits out of thin air - holds, and is relevant in a system where central banks target/control interest rates. Bottom Line: When central banks opt to control short-term interest rates, they must provide commercial banks with as much liquidity as the latter demands. In such a case, RRs and the RR ratio become almost irrelevant. Therefore, in an interest rate targeting system, banks' ability to originate loans/create money and deposits is not contingent on their reserves at the central bank. This point is greatly relevant to China. The PBoC: Shifting From Money To Interest Rate Targeting For the past few years, China’s central bank has silently moved away from controlling money growth to targeting interest rates. As a result, nowadays the PBoC has very little quantitative control over money/credit creation by commercial banks or the money multiplier. It is Chinese commercial banks that effectively drive money/credit/deposit creation. Chart I-3SHIBOR Crises In 2013 Forced PBoC ##br##To Start Targeting Interest Rates bca.ems_sr_2016_11_23_s1_c3 bca.ems_sr_2016_11_23_s1_c3 We suspect this shift in China's monetary policy management has been occurring since early 2014 on the heels of the so-called SHIBOR crisis, which erupted in June 2013 when interbank rates surged and was followed by another spike in interbank rates in December 2013 (Chart I-3). During these episodes, the PBoC enforced reserve requirements and thus did not provide liquidity to banks that were running short on it. In essence, it did whatever a central bank targeting money growth via control over RR would do. However, as interbank rates surged and banks complained, policymakers backed off, and provided banks with as much liquidity as they demanded. This stabilized interbank rates and, importantly, appears to have marked the PBoC's shift toward interest rate targeting. Thus, by de facto moving to a monetary system of targeting interest rates, the PBoC cannot effectively reinforce reserve requirements because it must supply any amount of reserves that commercial banks require to preclude a major spike in interbank rates. A few points illustrate that in fact the PBoC has been targeting short-term money market rates, and banks have expanded loans enormously despite their excess reserves being flat: Volatility in interbank rates has dropped substantially (Chart I-4), as the PBoC's claims on commercial banks has exploded 5.5-fold since the early 2014. Even though commercial banks' excess reserves have been flat, their lending has been booming - i.e., the money/credit multiplier has been rising (Chart I-5). This is only possible when the PBoC has been supplying reserves "on demand" or when it cuts the RR ratio. Since the RR ratio has not been cut over the past two years, it means that the former is true. Chart I-4Interbank Rate Volatility Has Fallen As ##br##PBoC Injected A Lot Of Liquidity bca.ems_sr_2016_11_23_s1_c4 bca.ems_sr_2016_11_23_s1_c4 Chart I-5China's Money/Credit Multiplier##br## Has Been Rising bca.ems_sr_2016_11_23_s1_c5 bca.ems_sr_2016_11_23_s1_c5 Just like central banks in advanced economies, the only way the PBoC can alter money/credit growth is if it lifts or cuts its interest rate target. Barring any changes to its policy rate, commercial banks, not the PBoC, determine money/loan/deposit creation in China. As to other factors that determine the amount of credit/money creation by commercial banks in China, we elaborated on these in the above-mentioned report. Bottom Line: It appears the PBoC has shifted toward targeting interest rates. Consequently, the PBoC cannot pretend to control money/credit origination unless it changes its interest rate target. Moreover, we reiterate that China's abnormal credit growth has been the result of speculative behavior among Chinese banks and borrowers, and not the natural result of the country's high savings rate. Oversupply Of RMBs = A Lower Currency As China's central bank has been printing RMBs and commercial banks have been "multiplying" them at a high rate (by originating loans), the supply of RMBs has continued to explode. Such an oversupply of local currency will continue to depress the value of the nation's exchange rate. The PBoC's liquidity injections have exploded in recent years (Chart I-6). The central bank has not only been offsetting the liquidity withdrawal due to its currency foreign exchange market interventions, but it has also been providing banks with as much liquidity as they require. The objective seems to have been to avoid a rise in interbank rates when corporate leverage is extremely high and banks are overextended. Since February 2015, the PBoC's international reserves have dropped by US$0.9 trillion, or 4.2 trillion RMB (Chart I-7). This means that the PBoC has withdrawn 4.2 trillion RMBs from the system. If the central bank did not re-inject these RMBs into the financial system, interbank rates would have skyrocketed. As the PBoC has injected RMBs into the system, it has effectively undone its RMB defense. The whole point of defending the exchange rate from falling or depreciating too fast is to shrink local currency liquidity. Yet, naturally, that would also lead to higher interbank rates. If the central bank chooses not to tolerate higher interest rates and continues to inject local currency into circulation, the RMB's depreciation will likely continue and accelerate. By injecting RMBs into the system, the monetary authorities have allowed banks to continue to lend, thereby creating enormous amounts of money and deposits. Banks create deposits when they lend. The Chinese banking system has a lot of deposits partially because commercial banks have lent too much. In short, the supply or quantity of money (RMBs) has continued to explode, despite massive capital outflows. Notably, if the PBoC did not lend RMBs to commercial banks, the latter's excess reserves would have plunged by 4 trillion RMB (Chart I-8) and banks would have been forced to pull-back their lending. Chart I-6PBoC's Liquidity Injections Have ##br##Exploded Since Early 2014 bca.ems_sr_2016_11_23_s1_c6 bca.ems_sr_2016_11_23_s1_c6 Chart I-7China: Foreign Exchange##br## Reserve Depletion bca.ems_sr_2016_11_23_s1_c7 bca.ems_sr_2016_11_23_s1_c7 Chart I-8China: What Would Have Banks' Excess Reserves##br## Been Without Borrowing From PBoC? bca.ems_sr_2016_11_23_s1_c8 bca.ems_sr_2016_11_23_s1_c8 Overall, in the current fiat money system, when a central bank targets interest rates, the monetary authorities can print unlimited high-powered money (bank reserves) and commercial banks can multiply it by creating enormous amounts of loans/deposits.3 However, there is no free lunch - no country can print its way to prosperity (otherwise all countries would have been very rich already). The negative ramifications of unlimited money creation are numerous, but this report focuses on the exchange rate implications. The growing supply of RMBs will lead to a much further drop in China's exchange rate. It seems Chinese retail investors and companies intuitively sense this, and are eager to get rid of their RMBs. This also explains Chinese investors' desire to overpay for any real or financial asset, domestically or abroad. We expect growing downward pressure on the RMB as capital outflows accelerate anew. Although China’s foreign exchange reserves are enormous in absolute U.S. dollar terms, they are low relative to money supply (Chart 9). The ratio of the central bank’s international reserves-to-broad money is 15% in China and it is relatively low compared with other countries (Chart 10). Chart I-9China: International Reserves Are Not##br## High Relative To Broad Money bca.ems_sr_2016_11_23_s1_c9 bca.ems_sr_2016_11_23_s1_c9 Chart I-10International Reserves-To-Broad##br## Money Ratio China's Money Creation Redux And The RMB China's Money Creation Redux And The RMB As a final note, the oversupply of local currency has not created inflation in the real economy because of massive overcapacity following years of booming capital spending. However, continued money creation will eventually lead to higher inflation. This does not seem imminent but we will be monitoring these dynamics carefully going forward. Bottom Line: China's banks have created too much RMBs and the PBoC has accommodated them. Such enormous supply of RMBs and mainland households' and companies' desire to get rid of their RMBs will lead to further yuan depreciation. Investment Implications: A Free-Fall For RMB And Asian Currencies The RMB's value versus the U.S. dollar will drop much further. Our new target range for US$/CNY is 7.8-8 over the next 12 months, or 11-14% below today's level. The forward market is discounting only 2.8% depreciation in the next 12 months (Chart I-11). We maintain our short RMB / long U.S. dollar trade (via 12-month NDF). A persistent relapse in the RMB's value will drag down other Asian currencies. In particular, the Korean won and the Taiwanese dollar have failed to break above important technical levels (their long-term moving averages), and have lately relapsed (Chart I-12). Chart I-11RMB Will Depreciate Much More##br## Than Priced In By Forwards RMB Will Depreciate Much More Than Priced In By Forwards RMB Will Depreciate Much More Than Priced In By Forwards Chart I-12Asian Currencies:##br##More Downside Ahead bca.ems_sr_2016_11_23_s1_c12 bca.ems_sr_2016_11_23_s1_c12 For the Korean won, we believe there is considerable downside from current levels. Consistently, we recommended shorting the KRW versus the THB trade on October 19.4 Chart I-13EM ex-China Currencies Total Return##br## (Including Carry): Is The Rally Over? bca.ems_sr_2016_11_23_s1_c13 bca.ems_sr_2016_11_23_s1_c13 Traders who believe in continued U.S. dollar strength, like we do, should consider shorting the KRW versus the U.S. dollar outright. For DM currencies, this means that the drop in the JPY has further to go. In emerging Asia, we are also shorting the MYR and the IDR versus the U.S. dollar and also versus Eastern European currencies such as the ruble and the HUF, respectively. As emerging Asian currencies depreciate versus the U.S. dollar, other EM currencies will likely follow. It is hard to see the RMB and other Asian currencies plunging and the rest of EM doing well. The total return (including the carry) of the aggregate EM ex-China exchange rate versus the U.S. dollar (equity market-cap weighted index) has failed to break above a critical long-term technical resistance, and has rolled over (Chart I-13). This is a bearish technical signal, implying considerable downside from these levels. As such, we maintain our core short positions in the following EM currencies outside Asia: TRY, ZAR, BRL and CLP and add COP to this list today. This is based on an assumption of diminished foreign inflows to EM and lower commodities prices. Arthur Budaghyan, Senior Vice President Emerging Markets Strategy & Frontier Markets Strategy arthurb@bcaresearch.com Andrija Vesic, Research Assistant andrijav@bcaresearch.com Colombia: Headed Toward Recession In our May 4 Special Report on Colombia,5 we argued that despite a bright structural backdrop this Andean economy was headed for a growth recession (i.e. very weak but still positive growth). Domestic demand has buckled and now we believe the nation could be on the verge of its first genuine recession in two decades (Chart II-1). Colombia's Achilles heel is its low domestic savings rate, reflected by a still large current account deficit financed by FDI and portfolio capital inflows (Chart II-2). As a result, low oil prices and rising global interest rates have exposed the nation's main cyclical vulnerability. Given the trade deficit is still large (Chart II-3) and our bias is that oil prices will be flat-to-down, a further retrenchment in domestic demand is unavoidable. Chart II-1Colombia's First Recession##br## In 20 Years? bca.ems_sr_2016_11_23_s2_c1 bca.ems_sr_2016_11_23_s2_c1 Chart II-2Colombia's Lingering Balance Of ##br##Payments Vulnerability bca.ems_sr_2016_11_23_s2_c2 bca.ems_sr_2016_11_23_s2_c2 Chart II-3A Weaker COP Will Force The ##br##Necessary Adjustment bca.ems_sr_2016_11_23_s2_c3 bca.ems_sr_2016_11_23_s2_c3 Going forward, the external funding constraint will continue to bite. Moreover, policymakers are trapped and will be unable to prevent growth from contracting. The central bank is stuck between the proverbial rock and hard place. Cutting interest rates will undermine the appeal of the peso to foreign investors. Raising rates to prop up the currency, however, will exacerbate the economy's downward momentum. In the end, downward pressure on the exchange rate and still high inflation mean the central bank will not cut rates soon (Chart II-4). Tight monetary policy in turn means that private sector credit will decelerate much more (Chart II-5). Chart II-4High (Well Above Target) Inflation Limits##br## Central Bank's Ability To Ease bca.ems_sr_2016_11_23_s2_c4 bca.ems_sr_2016_11_23_s2_c4 Chart II-5Colombia: Credit Growth Is ##br##Headed Much Lower bca.ems_sr_2016_11_23_s2_c5 bca.ems_sr_2016_11_23_s2_c5 Our marginal propensity to consume proxy, an excellent leading indicator for household spending, signals consumption is set to weaken even further (Chart II-6). Facing weakening demand, investment is set to continue contracting (Chart II-7) and, ultimately, unemployment will be much higher, reinforcing the downtrend in consumer expenditures. Chart II-6Colombian Domestic Demand##br## To Retrench Further bca.ems_sr_2016_11_23_s2_c6 bca.ems_sr_2016_11_23_s2_c6 Chart II-7Contracting Investment Bodes ##br##Poorly For Employment bca.ems_sr_2016_11_23_s2_c7 bca.ems_sr_2016_11_23_s2_c7 Meanwhile, fiscal policy will remain tight as Colombia's orthodox policymakers struggle to adjust the fiscal accounts to the structurally negative terms-of-trade shock in this oil-dependent economy. The current fiscal reform effort is very positive for sustainable long-run dynamics, as influential central bank board members have highlighted.6 Yet particular parts of the reform, such as raising VAT taxes from 16% to 19%, will almost inevitably lead to a drop in consumer demand. Furthermore, nominal government revenues are already contracting and a slumping economy means that the total fiscal effort will need to be greater than currently envisioned. Overall, with monetary and fiscal policy stimulus hamstrung by the nation's low domestic savings rate (i.e. large current account deficit), a mild recession seems very likely. And while a lot of weakness has already been priced into the nation's financial markets, we think there is still more downside ahead. For instance, the Colombian peso may be cheap in real (inflation-adjusted) terms, but it is highly vulnerable due to the nation's still wide current account deficit. This week we recommend re-instating a short position in the peso; this time against an equal-weighted basket of the U.S. dollar and the Russian ruble.7 Turning to equities, Colombian stocks have fallen sharply since 2014, mostly a reflection of the collapse of the nation's energy plays. At present bank stocks account for 60% this nation's MSCI market cap, and though we believe they will fare better than many other EM banking systems,8 they will not go unscathed by a recession. Still, orthodox policymaking should limit the downside in the performance of this bourse and sovereign credit (U.S. dollar bonds) relative to their respective EM benchmarks. Meanwhile, fixed-income investors should continue to bet on yield curve flattening by paying 1-year/ receiving 10-year interest rate swaps, a trade we have recommended since September 16, 2015.9 The recent steepening in the yield curve will prove unsustainable as the economy tanks. Bottom Line: Colombia is probably headed toward recession and policymakers are straightjacketed and cannot ease monetary and fiscal policies to prevent it. As such, the currency will be the main release valve and it will depreciate further. Go short the COP versus an equal-weighted basket the U.S. dollar and the Russian ruble. Dedicated EM equity and credit investors should maintain a neutral allocation to Colombia within their respective EM benchmarks. Continue to bet on flattening in the yield curve by paying 1-year/ receiving 10-year interest rate swaps. Santiago E. Gomez Associate Vice President santiago@bcaresearch.com 1 Please refer to the Emerging Markets Strategy Special Report, titled "Misconceptions About China's Credit Excesses", dated October 26, 2016. 2 Please refer to the Emerging Markets Strategy Special Report, titled "Misconceptions About China's Credit Excesses," dated October 26, 2016. 3 As we argued in Emerging Markets Strategy Special Report, titled "Misconceptions About China's Credit Excesses", dated October 26, 2016, it is new loans that create new deposits and vice versa. 4 Please refer to the section on Thailand in our Emerging Markets Strategy Weekly Report, titled " The EM Rally: Running Out Of Steam?" dated October 19, 2016. 5 Please refer to the Emerging Markets Special Report titled, "Colombia: A Cyclical Downturn Amid Structural Strength," dated May 4, 2016, available at ems.bcaresearch.com 6 Please see Cano, Carlos Gustavo "Monetary Policy in Colombia: Main Challenges 2016 -2017" Bank of America Merrill Lynch, Small Talks Symposium, October 7, 2016, Washington DC http://www.banrep.gov.co/sites/default/files/publicaciones/archivos/cgc_oct_2016.pdf 7 For more on the ruble please refer to the section on Russia in our Emerging Markets Weekly Report, dated November 16, 2016, titled, "Russia: Overweight Equities; Reinstate Long RUB / Short MYR Trade". 8 Please refer to the Emerging Markets Special Report titled, "Colombia: A Cyclical Downturn Amid Structural Strength" dated May 4, 2016, available at ems.bcaresearch.com 9 Please refer to the section on Colombia in our Emerging Markets Weekly Report, dated September 15, 2015, titled "Colombia: An Incomplete Adjustment", available at ems.bcareseach.com Equity Recommendations Fixed-Income, Credit And Currency Recommendations

Highlights Most narratives surrounding G7 bond yields, the U.S. dollar, Chinese credit/fiscal impulses, and the RMB exchange rate - which justified the EM rally from February's lows - have been overturned. To be consistent, this warrants a relapse in EM risk assets. In China, recent property market and marginal credit policy tightening will weigh on growth. Feature The more recent strength in Chinese and emerging markets' (EM) manufacturing PMI indexes as well as the bounce in industrial metals prices have gone against our negative view on EM/China growth and related markets. While it is hard to predict market patterns over the next several weeks, we maintain that the EM rally is on borrowed time, and that the risk-reward profile for EM risk assets (stocks, credit markets and currencies) remains very unfavorable. Tracking Correlations And Indicators The overwhelming majority of indicators and variables that supported the rally in EM since February have reversed in recent months. Specifically: China's credit and fiscal spending impulses have rolled over (Charts I-1 and Chart I-2, on page 1). This will likely lead to a rollover in mainland industrial activity early next year (Chart 1, top panel). Similarly, this bodes ill for much-followed Chinese ex-factory producer prices - i.e., producer price deflation will probably recommence early next year (Chart I-1, bottom panel). Chart I-1China: Industrial Sectors To Retreat? bca.ems_wr_2016_11_09_s1_c1 bca.ems_wr_2016_11_09_s1_c1 Chart I-2China: Credit And Fiscal Impulses China: Credit And Fiscal Impulses China: Credit And Fiscal Impulses In a nutshell, the strong credit and fiscal impulses of late 2015 and early 2016 explain the stabilization and mild improvement in the Chinese economy during the past few months. However, these same impulses project renewed weakness/rollover in the economy in early 2017. If financial markets are forward looking, they should begin pricing-in deteriorating growth momentum sooner than later - especially as Chinese policymakers are announcing marginal tightening policies (see below for more details). One of the narratives that triggered the EM and global equity rally in February was speculation that there was a "Shanghai accord" between global central banks. According to this narrative, the People's Bank of China (PBoC) promised not to devalue the RMB in exchange for the Federal Reserve not hiking rates. Since then, the RMB has continued to depreciate, both versus the greenback and the CFETS1 basket. Yet EM and global stocks have completely disregarded the RMB depreciation (Chart I-3). We do not have good explanation as to why. Indeed, the RMB has weakened meaningfully, despite the PBoC's massive currency defense: the latter's foreign exchange reserves have shrunk further since then (Chart I-4), as capital flight has exceeded the enormous current account surplus by a large margin. Chart I-3Investors Are ##br##Complacent About RMB bca.ems_wr_2016_11_09_s1_c3 bca.ems_wr_2016_11_09_s1_c3 Chart I-4China: Foreign Exchange ##br##Reserves Still Shrinking bca.ems_wr_2016_11_09_s1_c4 bca.ems_wr_2016_11_09_s1_c4 Chart I-5PBoC Liquidity Injections ##br##Have Been Enormous bca.ems_wr_2016_11_09_s1_c5 bca.ems_wr_2016_11_09_s1_c5 The PBoC's selling of U.S. dollars to prop up the yuan has drained domestic currency liquidity and one would expect interbank rates to rise. However, the PBoC has been re-injecting RMBs into the system to keep interest rates low (Chart I-5). Such RMB liquidity proliferation makes further declines in the currency's value all the more likely. We expect the RMB to continue depreciating. Yet global financial markets have become extremely complacent about the potential for additional RMB depreciation. After having been bullish on U.S./G7 bonds for the past several years, in our July 13 Weekly Report,2 we highlighted that U.S./G7 bond yields would rise and closed our strategic short EM equities/long 30-year U.S. Treasurys position. Even though U.S./G7 bond yields have risen since July, EM equities have not declined. Given that falling G7 bond yields were used as justification for the EM rally, the opposite should also hold true. We expect U.S. bond yields to rise further. Our EM Corporate Health Monitor - constructed using bottom-up financial variables of companies with outstanding U.S. dollar corporate bonds - points to a reversal in the EM corporate credit market rally (Chart I-6). Furthermore, EM sovereign and corporate credit spreads have tightened considerably and are now very overbought and expensive. As we argued in our Special Report titled EM Corporate Health Is Flashing Red3 that introduced the EM Corporate Financial Health (CFH) Monitor, EM corporate credit spreads are as expensive as they were before they began widening in 2013 and 2014 (Chart I-7). Chart I-6EM Corporate Bond Rally To Reverse? EM Corporate Bond Rally To Reverse? EM Corporate Bond Rally To Reverse? Chart I-7EM Corporate Spreads Are Too Tight EM Corporate Spreads Are Too Tight EM Corporate Spreads Are Too Tight Finally, the U.S. dollar sold off early this year, but it has held firm in recent months. Nevertheless, EM risk assets have not retreated, despite the greenback's strength (Chart I-8). Few would argue that sharp U.S. dollar appreciation is negative for EM risk assets, but there is a debate among investors and analysts about whether EM risk assets can rally amidst a gradual appreciation in the U.S. dollar. Turning to the empirical evidence, Chart I-9 reveals that in the past 30 years any U.S. dollar appreciation - whether gradual or not - even versus DM currencies has coincided with weakness in EM share prices. Chart I-8EM Investors Have ##br##Ignored U.S. Dollar Strength bca.ems_wr_2016_11_09_s1_c8 bca.ems_wr_2016_11_09_s1_c8 Chart I-9EM Equities And ##br##U.S. Dollar: A 30 Year History EM Equities And U.S. Dollar: A 30 Year History EM Equities And U.S. Dollar: A 30 Year History Bottom Line: The majority of narratives that justified the EM rally from February's lows have been overturned. To be consistent, this warrants a relapse in EM risk assets. China's Credit And Property Tightening In recent weeks, there have been numerous policy tightening efforts in China. In particular: At the annual World Bank/IMF meetings in Washington last month, PBoC Governor Zhou Xiaochuan stated that once markets stabilized there would no longer be additional large increases in bank credit. His exact words were: "With the gradual recovery of the global economy, China will control its credit growth".4 As U.S. and European PMIs have firmed up and U.S. employment and wage growth is robust, Chinese policymakers will be emboldened to moderate unsustainable credit growth and not to repeat the massive fiscal push of early this year. In a bid to curb excessive bank credit growth and discourage "window dressing" accounting, the PBoC announced on October 255 that going forward it will include off-balance-sheet wealth management products (WMPs) in the calculation of banks' quarterly Marco Prudential Assessment ratios, starting from the third quarter. The clampdown on WMP accounting will reduce banks' capital adequacy ratios (CARs). One key reason that banks had aggressively boosted the size of their off-balance-sheet WMP assets was that they were not required to have capital charges against them, helping banks extend more credit while complying with CARs. In short, Chinese banks' CARs are inflated. This policy measure along with provisioning and writing-off non-performing loans, if reinforced, could meaningfully reduce the CARs of all Chinese banks, especially small- and medium-sized ones, as well as force them to reduce the pace of credit expansion. Given that the majority of medium and small banks have been more aggressive than the country's five biggest banks in expanding credit in recent years, this may have a damping effect on credit growth in 2017. In fact, the 110 medium and small banks retain 60% of on- and off-balance-sheet credit claims on companies, while the five largest banks hold 40% (Table I-1). Hence, credit trends in small and medium banks are at least as important as those among large banks. Table I-1China: Five Largest Banks Hold Only 40% Of Credit Assets EM: Defying Gravity? EM: Defying Gravity? Finally, a number of cities have announced various tightening measures on property markets of late, including the re-launch of house purchasing restrictions and increases in minimum down payments. Similar restrictions on home purchases served as an efficient tool for curbing property purchases in 2013-14, and there is no reason why it will be different this time around. This is especially true given the market is more expensive than it was back in 2013. In addition, the government has curbed financing for property developers. The biggest economic risk remains construction activity. Even though housing sales and prices have skyrocketed by 20-40% in the past 12 months (Chart I-10, top and middle panels), residential floor space started has been very timid - it has in fact failed to recover (Chart I-10, bottom panel). As residential property sales contract again due to new purchasing restrictions, property developers will certainly curtail new investment, and housing construction activity will shrink anew. The same is true for commercial properties (Chart I-11). Chart I-10China's Residential Market: ##br##Demand, Prices And Starts China's Residential Market: Demand, Prices And Starts China's Residential Market: Demand, Prices And Starts Chart I-11China's Non-Residential ##br##Market: Demand And Starts China's Non-Residential Market: Demand And Starts China's Non-Residential Market: Demand And Starts An interesting question is why property starts have been so weak, as indicated in the bottom panels of Chart I-10 and Chart I-11 - particularly when both floor space sold (units) and property prices have surged exponentially in the past 12 months. Our view is that there is a large hidden inventory overhang in the Chinese property market. For example, government data on residential floor space started, completed and under construction attest that there is still a large gap between floor space started versus completed (Chart I-12). From these data/charts and the enormous leverage carried by property developers, we infer the latter have been accumulating / carrying on their balance sheets vast amounts of inventory in excess of what market-based sources suggest, and what is widely followed by analysts. It is very hard to make sense of the Chinese property inventory data, but we suspect these market-based data sources may track only inventories that have been completed and released to the market - and do not account for inventories classified as "under construction". For residential housing, according to government data the "under construction floor space" is 5 billion square meters (Chart I-13, top panel), which is equal to 3.5-4 years of sales at the fervent pace of the past 12 months (Chart I-13, bottom panel). Another way to assess this is as follows: Assuming an average construction cycle of three years, there will be supply of new housing in amounts of 16.7 units in each of the next three years. This compares with sales of 13.3 million units in the past 12 months that occurred amid a buying frenzy and booming mortgage lending. Faced with a potential drop in sales due to the recent purchasing restrictions, elevated inventories, enormous leverage (Chart I-14), and tighter financing, property developers will most likely curtail new starts. In turn, a reduction in property starts means less construction activity next year, and weak demand for commodities. Consistent with the rollover in the fiscal spending impulse, infrastructure spending will likely also lose its potency in early 2017. Chart I-12China's Residential ##br##Market: Hidden Inventories bca.ems_wr_2016_11_09_s1_c12 bca.ems_wr_2016_11_09_s1_c12 Chart I-13Chinese Real Estate: Massive ##br##Volumes Under Construction Chinese Real Estate: Massive Volumes Under Construction Chinese Real Estate: Massive Volumes Under Construction Chart I-14Leverage Of Chinese ##br##Listed Property Developers bca.ems_wr_2016_11_09_s1_c14 bca.ems_wr_2016_11_09_s1_c14 Bottom Line: Recent property market and marginal credit policy tightening will weigh on construction activity and depress Chinese demand for commodities and industrial goods next year. Confirmation Bias, Or Bias Based On Fundamentals? Why did we not follow the indicators discussed above from February through June, when the EM rally emerged and these indicators bottomed? Do we have a confirmation bias? We did not recommend playing the EM rebound early this year because we did not believe the rally would last this long or go this far. If we had had conviction about the duration and magnitude of the rally, we would have changed our strategy - tactically upgrading EM risk assets despite our negative structural and cyclical views. Simply put, we were wrong on strategy. In our April 13, 2016 Weekly Report,6 we argued that based on China's injection of massive amounts of fiscal and credit stimulus, growth would marginally improve in the months ahead. Yet, we stopped short of recommending chasing the EM rally given the menace of numerous cyclical and structural negatives surrounding the EM/China growth outlook. As to the reasons why we put more emphasis on some indicators and less on others at various times, we have the following points: We are biased in so far as our assessment and analysis of EM/China is based on fundamentals. In this sense, we are biased towards centering our investment strategy on fundamentals. Specifically, given our view/analysis that EM/China have credit bubbles/excesses, rapidly falling or weak productivity growth and record-low return on capital (Chart I-15), we cannot help but to have a fundamentally bearish bias on EM. This, in turn, means that we view any rally in EM risk assets or uptick in EM/China economic indicators with suspicion and likely as unsustainable. The opposite also holds true. All in all, if we are wrong on our fundamental view and analysis, we will be wrong on financial markets. When investors expect a bear market, they are better off selling rallies and not buying dips. When an asset class is in a multiyear bull market, it pays off to buy dips rather not sell rallies. Unless one can time market swings well, it is hard to make money on the long sides of bear markets. Similarly, it is difficult to profit from short positions in bull markets. In brief, countertrend moves are about timing. Timing does not depend on fundamentals. It is often a coin toss. Typically we do not recommend clients invest based on a coin toss. For example, it is impossible to rationalize why the EM rally did not begin following the August 2015 selloff, but instead started in February 2016. In late August 2015, with carnage in EM risk assets pervasive, it was clear that Chinese policymakers would stimulate and in fact the massive fiscal stimulus was initiated in August/September 2015 not in 2016. Similarly, China's manufacturing PMI bottomed in September 2015, not in 2016 (Chart I-16). Chart I-15EM Non-Financial Return ##br##On Equity Is At All Time Low EM Non-Financial Return On Equity Is At All Time Low EM Non-Financial Return On Equity Is At All Time Low Chart I-16China's Manufacturing PMI ##br##Bottomed In October 2015 China's Manufacturing PMI Bottomed In October 2015 China's Manufacturing PMI Bottomed In October 2015 In September 2015, EM and global equities rebounded, but chasing momentum at that time did not pay off as risk assets cratered in the following months. This is all to say that timing markets is often a random walk. We do attempt to time market moves that go along with our fundamental bias, but prefer not to time market moves that go against the primary trend. We assume any countertrend move is typically short-lived and unsustainable. That said, we also realize these moves can be very painful for investors if they last long enough, like this EM rally. Finally, we often get questions on fund flows. We do not make investment recommendations based on fund flows - even though we recognize they are very important in driving markets. The reason is that there is no comprehensive data on global fund flows that one can analyze and make reasonably educated bets. The often-cited EPRF dataset only tracks inflows and outflows of mutual funds and ETFs. It does not account for flows and positioning of various asset managers, sovereign funds, pension funds, insurance companies, hedge funds and private wealth managers, among many others. What's more, the EPRF dataset only covers the funds located in advanced countries and offshore jurisdictions, but not emerging countries where investment pools have become large and important. In brief, the available investment flow and portfolio positioning data are not comprehensive at all, and they cannot be relied upon too much to make investment recommendations. In this vein, a question arises: Why can't flows into EM sustain the current rally for a while even though it is not based on fundamentals? In this context, let's consider the case of the rally in euro area share prices when markets sensed the arrival of the European Central Bank's quantitative easing efforts at the beginning of 2015. There was a fervent rush to buy/overweight euro area stocks heading into the QE announcement by the ECB. European bourses surged. Nevertheless, euro area equity prices have been sliding and massively underperforming the global equity benchmark since March 2015 (Chart I-17). The reason the ECB's QE has not helped euro area stocks is because their fundamentals were bad - profits have been shrinking despite the ECB's QE. We suspect EM stocks and currencies will have a similar destiny: EM profits will disappoint considerably, and the current rally will prove unsustainable. Notably, net EPS revisions have so far failed to move into the positive territory (Chart I-18). Chart I-17Euro Area Stocks And EPS: ##br##Why The QE Rally Proved To Be Bogus Euro Area Stocks And EPS: Why The QE Rally Proved To Be Bogus Euro Area Stocks And EPS: Why The QE Rally Proved To Be Bogus Chart I-18EM Stocks And EPS: ##br##Earning Revisions Are Still Contracting bca.ems_wr_2016_11_09_s1_c18 bca.ems_wr_2016_11_09_s1_c18 Arthur Budaghyan, Senior Vice President Emerging Markets Strategy & Frontier Markets Strategy arthurb@bcaresearch.com 1 China Foreign Exchange Trading System. 2 Please refer to the Emerging Markets Strategy Weekly Report, titled "Risks To Our Negative EM View," dated July 13, 2016; a link is available on page 15. 3 Please refer to the Emerging Markets Strategy Special Report, titled "EM Corporate Health Is Flashing Red," dated September 14, 2016; a link is available on page 15. 4 Please see http://www.pbc.gov.cn/goutongjiaoliu/113456/113469/3155686/index.html 5 Please see http://www.pbc.gov.cn/goutongjiaoliu/113456/113469/3183204/index.html 6 Please refer to the Emerging Markets Strategy Weekly Report titled, "Revisiting China's Fiscal And Credit Impulses," dated April 13, 2016; a link is available on page 15. Equity Recommendations Fixed-Income, Credit And Currency Recommendations