Economy

Special Report

Highlights The CAD and AUD have tactical upside; however, this may well prove to be the last hurrah before some serious declines play out. This time domestic – not global – factors will drive the CAD and AUD lower. Canada and Australia are hitting the end game for their respective debt supercycles as rising U.S. rates will lift the global cost of capital. Canadian and Australian house prices and debt loads are too elevated; a reversal of these excesses is likely to push these two countries toward liquidity traps. These liquidity traps will cause the R-star in Canada and Australia to fall, lagging well behind the U.S. Canada and Australia are uncompetitive, suggesting external demand will not come to their respective rescue, at least not until after the CAD and AUD have fallen significantly. The CAD may fall first, but the AUD has more downside ultimately; not only is Australia even less competitive than Canada, but the Aussie is also more expensive than the Loonie. Feature The Canadian and Australian dollars are in the process of rebounding. This is not surprising. By the end of 2018, both these currencies were deeply oversold, and the recent easing in global financial conditions, helped by the Federal Reserve’s pause, is fueling their rebound (Chart 1). Moreover, pessimism toward China has hit an extreme, yet Sino-U.S. trade relations seem on the cusp of improving and Chinese policymakers are increasingly trying to manage the downside in the Chinese economy. This setup is normally supportive for the Canadian and Australian dollars (Chart 2). Chart 1Financial Conditions Point To A Tactical Rebound In The AUD And The CAD...   Chart 2...So Does Chinese Reflation While we have been recommending that our more tactically minded clients play this rally,1  the longer-term outlook for the CAD and AUD remains poor. These countries are getting closer to the end of their respective debt supercycles. Consequently, the CAD and AUD need to trade at much larger discounts to fair value in order to be attractive. Way Too Much Debt Canada and Australia have become victims of their own success. Canada and Australia have seen real estate prices rise for more than two decades. At first, rising prices reflected solid valuations, growing populations and rising prosperity. However, things changed around the Great Financial Crisis. During this traumatic event, the Bank of Canada and the Reserve Bank of Australia both dropped interest rates by 4.25%. Since both countries’ banking sectors escaped the crisis unscathed, and households did not experience similar losses of wealth as those in the U.S., Ireland or Spain, credit growth remained strong. A real estate bubble became the natural consequence of this easy monetary policy. Banks pushed credit to households, and households – impressed by the solid performance of real estate prices, attracted by low interest rates, and enamored with the dream of easy riches – willingly took on mortgages and piled into the property market. A feedback loop ensued, whereby rising collateral values made credit even easier to access, fomenting further house price gains and even-easier credit conditions. Today, we stand at the end of this process. Vancouver and Toronto in Canada, and Sydney and Melbourne in Australia are some of the most expensive real estate markets in world in terms of price-to-income ratios, when one controls for population density (Chart 3). This has created major systemic risks for both countries. Few would care about the systemic risk created by elevated house prices if debt loads were small. However, in both countries, household indebtedness makes Americans circa 2007 look like a frugal bunch. In Canada, household debt has now reached 176% of disposable income, or 100% of GDP, while in Australia, the same ratios are 189% and 121%, respectively. This is well above the levels that prevailed in the U.S. in 2007 (Chart 4). Mortgage debt alone represents 108% and 140% of disposable income in Canada and Australia, respectively. Moreover, Canadian and Australian households also spend 14.5% and 15.6% of their incomes servicing debt, which also compares unfavorably with the U.S. in 2007. Chart 4ACanadians And Australians Make Americans Look Frugal Chart 4BCanadians And Australians Make Americans Look Frugal Canadian and Australian households thus seem close to having reached their maximum debt loads. Moreover, measures taken in Canada and Australia to limit foreign money inflows and constrain bank lending are beginning to bite. In both countries, real estate transactions are slowing, with property sales declining by 20% and 8% in Canada and Australia, respectively. House prices too are being hit. House prices in Vancouver and Toronto peaked by 2018, and in Sydney and Melbourne in 2017. Residential construction is likely to be the first victim. Real estate inventories in both these countries have been rising, courtesy of the frenetic pace of housing starts going on for decades. Today, residential investment represents 7% of GDP in Canada and 5% of GDP in Australia (Chart 5). Thus, slowing real estate activity could curtail Canadian and Australian GDP by 2% if we move back to the real estate environment that prevailed in the mid-1990s. This would also imply large hits to employment as construction, real estate and finance have created 336-thousand and 250-thousand jobs in Canada and Australia since 2009, respectively. Chart 5AA Decline In Construction Activity Would Be A Vicious Hit To Canada And Australia (1) Chart 5BA Decline In Construction Activity Would Be A Vicious Hit To Canada And Australia (2) Consumption too is likely to suffer. Without a growing wealth effect and with declining equity in their houses, Canadian and Australian households are likely to curtail consumption – consumption that has contributed 60% and 30% of Canada’s and Australia’s cumulative GDP growth since 2009. Already, we are seeing slowing Canadian and Australian retail sales – right behind drops in housing activity. The biggest and most dangerous risk is that Canada and Australia teeter on the verge of falling into a liquidity trap, like the U.S. after 2007. As Chart 6 illustrates, propelled by households binging on cheap money in the form of mortgages, Canadian and Australian banks have managed to maintain higher levels of return on equity after the financial crisis. This robust profitability will decline if non-performing loans, which so far remain low, grow in response to weakening house prices and fragile household financial health (Chart 7). Chart 6Canadian And Australian Banks Remain Profitable... Chart 7...As Long As NPLs Do Not Rise Rising NPLs and declining RoEs tend to limit the willingness of banks to lend. Just as crucially, the poor health of households and falling real estate prices is likely to also limit demand for credit. This combination was behind the sharp decline in the U.S. money multiplier in 2008. No matter how much reserves the Federal Reserve would inject in commercial banks via QE programs, broader money would not respond. A similar fate is likely to ensue in Canada and Australia (Chart 8). The velocity of money is also likely to fall if households are not willing to take on debt anymore and instead focus on rebuilding their financial buffers. Chart 8Canada And Australia Have Avoided A Liquidity Trap... So Far The consequence of this monetary constipation will be much lower interest rates. When an economy enters a liquidity trap, as was the case in the U.S. after 2007, in Japan since the 1990s, or in Europe after 2010, the neutral real rate of interest, the so-called R-star, falls to zero or even lower. Essentially, no matter how low interest rates fall, they cannot equilibrate the demand and supply for savings. Everyone wants to save, no one wants to borrow, and banks are unwilling to lend. This fate looks increasingly likely for both Canada and Australia over the coming two years. Bottom Line: The Canadian and Australian real estate markets have enjoyed incredible runs for more than two decades. Now, not only are real estate prices in these two nations very expensive, households have been left with prodigious debt loads. As real estate activity slows, residential construction will suffer, but most importantly, these two countries are likely to teeter toward becoming liquidity traps as banks curtail lending and households curtail borrowing. This will result in structurally lagging interest rates. Why Now? Betting on the end of the Canadian and Australian housing bubbles has so far been mugs games. Why is the situation different now? Because the U.S. economy is stronger. Until now, very low global interest rates have kept the Canadian and Australian housing bubbles afloat, but rising U.S. interest rates are now putting upward pressure on mortgage rates in both Canada and Australia (Chart 9). This simply reflects the fact that U.S. rates represent the ultimate opportunity cost of investing outside the international reserve currency, the U.S. dollar. After years of household deleveraging, the U.S. seems to be able to handle higher rates. However, because Canadian and Australian balance sheets are much weaker, their tolerance for higher rates is substantially lower. Chart 9Higher U.S. Rates Threaten Canadian And Australian Households BCA sees further upside for U.S. rates and thus for the global cost of capital. In other words, we do not anticipate the Fed’s pause to last beyond June. The following reasons underpin this view: The U.S. labor market is increasingly inflationary. The employment-to-population ratio for prime-age workers continues to rise, which historically has boosted labor costs (Chart 10). The New York City Fed Underlying Inflation Gauge points toward higher core inflation (Chart 11). Moreover, Ryan Swift argues in BCA’s U.S. Bond Strategy that an unfavorable base effect will dissipate after February, further reinforcing the upside risk to inflation.2  Being the only component of our Fed Monitor moving toward “easy money required” territory, the tightening in U.S. financial conditions last year was the lynchpin behind the Fed’s pause. The other components of the Fed Monitor have not deteriorated significantly, and they still argue in favor of further rate hikes (Chart 12). Thus, if the recent easing in financial conditions can persist, the Fed will hike again this year.   Chart 11Budding U.S. Inflationary Pressures   Chart 12The Fed Is Pausing Because Of Tightening Financial Conditions, Not The Economy Finally, U.S. productivity is set to pick up over the coming two years. Since a rising capital stock boosts productivity, the recent strength in capex augurs well (Chart 13). Moreover, the demand deficit created by the deleveraging of U.S. households has weighed on productivity. As U.S. credit growth picks up, so will productivity. This is important as rising productivity lifts the neutral rate, and thus creates more room for the Fed to lift interest rates. Chart 13Upside For U.S. Productivity Equals Upside For U.S. Rates Ultimately, all these factors point to higher U.S. rates. As such, it suggests that Canadian mortgage rates, and to a lesser extent Australian ones as well, will experience upward pressure – exactly at the time when households in these two countries are most vulnerable to higher rates. Bottom Line: Higher U.S. rates are the main reason why we expect the Canadian and Australian housing markets and economies to buckle now, finally heeding the call of doomsayers. Higher U.S. rates lift the global cost of capital. While U.S. households are in robust shape and therefore better able to handle higher rates, the same cannot be said about Canadian and Australian households. Can the External Sector Come To The Rescue? This is unlikely. After years of commodity booms and strong domestic demand supported by rising household wealth, the Canadian and Australian manufacturing sectors have been greatly diminished. Much capacity has vanished, and it will be difficult to replace the lost output from falling domestic demand by exports of manufactured goods. The Australian and especially the Canadian corporate sectors are also already heavily indebted, and thus, it could take quite some time before capacity is expanded. Complicating the situation, Canada and Australia are not competitive exporters anymore. As the top panel of Chart 14 shows, since 1980, U.S. unit labor costs have risen by 156%, but they have risen by 183% in Canada and by a stunning 282% in Australia. Productivity trends paint a similar, albeit less dramatic picture. Since 1980, U.S. labor productivity has risen 22% versus its trading partners; in Canada it has declined by 20%, and in Australia, by 5%. Consequently, both Canadian and Australia labor will have to cheapen. Historically, the mechanism through which labor costs decline is higher unemployment, which forces a painful adjustment in wages. These adjustments are likely to force both interest rates and currencies lower. Chart 14Canada And Australia Are Uncompetitive Could China come to the rescue? Via higher commodity prices, both Canada and Australia have been major beneficiaries of the Chinese economic boom. However, while China today is trying to contain its economic deceleration, Chinese policymakers remain fixated on controlling credit growth. This means that China is unlikely to go on another debt binge similar to what transpired in 2009 or in 2015-‘16. As a result, the recent uptick in commodity prices is unlikely to last long. More fundamentally, China is not only trying to move away from its debt-led growth model: It is also trying to move away from its investment-led growth model. This means that the commodity intensiveness of the Chinese economy is likely to decline. China’s emphasis on controlling air pollution will strengthen this trend. As Chart 15 illustrates, when the share of Capex as a percentage of Chinese GDP declines, so does the labor participation rate of Canada and Australia relative to the U.S. This decline in relative participation rates is associated with falling CAD and AUD values versus the U.S. dollar, a consequence of falling growth potential and interest rates. Chart 15AChanging Chinese Growth Model Points To Falling Canadian And Australian Participation Rates (1) Chart 15BChanging Chinese Growth Model Points To Falling Canadian And Australian Participation Rates (2) Bottom Line: Canada’s and Australia’s lack of manufacturing capacity, poor competitiveness, and China moving away from its investment-led growth model suggest that a deflationary environment will ultimately develop in these two nations, at least relative to the U.S. Moreover, the structurally negative outlook on consumption, debt growth and employment suggests that Canadian and Australian neutral rates are likely to fall relative to the U.S. These economic forces point to deeper lows this cycle in the CAD and AUD against the USD. Investment Implications Based on this economic backdrop, both the Canadian and Australian dollar could suffer significant downside in the coming years as their fair value is likely to fall, dragged by interest rates that will lag those in the U.S. However, if an asset is cheap enough, it may nonetheless be an attractive buy. The CAD and AUD do not fall into that camp. Today, the CAD trades in line with our long-term fair-value model, implying that if its fair value falls, the CAD provides zero insulation and will therefore also have to decline. The AUD is in an even worst spot as it currently trades above its fair value (Chart 16). Additionally, the Australian current account deficit is larger than Canada’s. Chart 16The CAD And AUD Are Not Cheap Enough To Compensate For Secular Risks In terms of timing, the Loonie could start weakening before the Aussie. The Canadian housing bubble is likely to collapse first as Canadian mortgage rates are more tightly linked to U.S. ones than Australian rates are. Moreover, the Canadian economy seems even more levered to rising real estate prices than that of Australia. However, a collapse in Vancouver and Toronto housing prices will promptly catalyze similar weaknesses in Sydney and Melbourne. Thus, while the CAD may be the first to take the great plunge, the AUD will not be far behind. Ultimately, the AUD will suffer the greatest decline. Obviously, the more onerous pricing of the AUD contributes to this assessment, but so does the greater lack of competitiveness in Australia than in Canada. Australia is likely to endure deeper deflationary pressures as its labor costs need greater adjustments. Furthermore, Australia already suffers from a larger degree of underutilized labor than Canada. Since the currency – not wages – is likely to withstand the bulk of the competiveness adjustment, this implies that the AUD has more work to do than the CAD. The more expensive valuations of Australian assets also handicap the Aussie versus the Loonie. Australian real estate is pricier than Canadian property, and Australian stocks are more expensive (Chart 17). This means that Australians could end up with deeper holes in their balance sheets than Canadians, and that Australia has scope to witness greater outflows of capital than Canada. Chart 17Canadian Financial Assets Are Cheaper Than Australian Ones... Where Australia shines relative to Canada is in terms of the ability of fiscal authorities to respond to an economic slowdown. Canadian public debt stands at 90% of GDP versus 41% of GDP in Australia. Canada’s cyclically-adjusted primary deficit is already deteriorating, while Australia’s is improving (Chart 18). This means that the Australian governments have deeper pockets and a greater capacity to support domestic demand than Canada’s. This could cushion the deflationary impact in Australia relative to Canada. That being said, the Japanese, Spanish or U.S. experiences argue that once a real estate bubble bursts, fiscal spending can cushion some of the pain, but it cannot eradicate the problem – at least not until banks are recapitalized and the private sector is once again ready to borrow, something that takes years of balance-sheet rebuilding. Chart 18...But Australia Has More Fiscal Space Bottom Line: Both the CAD and AUD are likely to experience substantial downside over the coming years. The CAD and AUD are not cheap enough to compensate for a BoC and RBA that will greatly lag the Fed. While the CAD may weaken first, the AUD will suffer more long-term downside. The Aussie is more expensive, Australia is less competitive than Canada, and it could suffer greater outflows of capital. Continue to underweight Australian and Canadian assets in global portfolios as the AUD and CAD will drag their performance down. Remain short AUD/CAD on a structural basis.   Mathieu Savary, Vice President Foreign Exchange Strategy mathieu@bcaresearch.com Footnotes 1 Please see Foreign Exchange Strategy Weekly Report, titled “Global Liquidity Trends Support The Dollar, But…”, dated January 25, 2019, 2018, available at fes.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, “Buy Corporate Credit”, dated January 15, 2019, available at usbs.bcaresearch.com Trades & Forecasts Forecast Summary Core Portfolio Tactical Trades Closed Trades

Feature Half Way Back Since BCA went overweight global equities in late December, the MSCI ACWI index has rallied by 8% and the S&P 500 is back to only 8% off its September historical high. So far, this has been little more than a technical rally from the extreme oversold position in Q4. But with U.S. economic growth still resilient, earnings likely to grow healthily again this year (albeit more slowly than in 2018), and the valuation of risk assets (both equities and credit) no longer a headwind, we expect the rally to continue for some time, and so reiterate our overweight on equities. Recommendations True, there have been some disappointments in U.S. data in recent weeks. In particular, the December manufacturing ISM fell sharply to 54.3 from 59.3, raising fears that the U.S. is starting to decelerate in line with other regions (Chart 1). But the ISM may have been affected by the government shutdown and, overall, U.S. data still look solid, with the Citigroup Economic Surprise Index beginning to rebound, and stronger than in other regions (Chart 2). The residential housing market, which was exhibiting signs of stress last year, with existing home sales -6.4% YoY in December, is showing the first signs of stabilization, helped by mortgage interest rates that are now 50 BPs off their recent peak (Chart 3). Chart 1How Worrying Is The U.S. Slowdown? Chart 2U.S. Data Surprisingly Positive Chart 3Housing Market Should Stabilize In particular, the outlook for consumption looks healthy, with average hourly earnings growing at 3.3% YoY, consumer confidence close to an historic high, and the savings rate above 6%. Unsurprisingly, then, retail sales have boomed in recent months (Chart 4). Unless consumer confidence is dented by a repetition of the government shutdown or some other shock, consumption (68% of GDP, remember) should grow strongly this year. Add to this a residual positive impact of close to 0.5% of GDP coming from last year’s fiscal stimulus, and it is hard to imagine the U.S. going into recession over the next 12 months. Chart 4Consumption Booming The Fed will probably go on hold for now, however, given the market jitters in Q4. We are likely back to a situation like that in 2015-2016, where the Fed Policy Feedback Loop becomes the key factor for markets (Chart 5). When financial conditions tighten, with stock prices falling and the dollar appreciating, the Fed turns more dovish. However, this triggers a rally in risk assets and loosens financial conditions, allowing the Fed to start hiking again. With the tightening in financial conditions over the past six months, the Fed is likely to err on the side of caution for now (Chart 6). However, if our macro view is correct – and as inflation starts to pick up again after April, partly due to the base effect – the Fed will want to continue withdrawing accommodation over the course of this year. The Fed Funds Rate, at around 2.4% is still two hikes below what the FOMC sees as the neutral level of interest rates (the 2.8% terminal rate in the FOMC dots). We see the Fed, therefore, raising rates in June and perhaps hiking two or even three times this year. By contrast, the futures market assigns only a 25% probability of even one rate hike this year, and is even pricing in a small probability of a cut. Chart 6Tighter Conditions Mean More Cautious Fed Clearly, there are plenty of risks to the scenario of growth continuing. But those in the hands of President Trump, especially the trade war with China and the fight over funding of the wall on the border with Mexico, we don’t see as being serious impediments. Trump is fully aware that he is unlikely to be reelected in November 2020 if the U.S. is in recession by then. Every incumbent U.S. president since World War Two who fought for reelection during a recession failed to be reelected (Chart 7). The view of BCA’s geopolitical strategists, therefore, is that the White House and Congressional Democrats will agree to concessions to end the shutdown before the end of the current three-week stop-gap period. Less likely, Trump will declare a national emergency that will cause much controversy but have little impact on the economy. Our strategists also argue that there is a 45% probability of trade negotiations with China producing a result (at least a short-term one the president can boast about) before the March 1 deadline, and a further 25% probability of the deadline being extended without further sanctions being imposed.1 Chart 7Trump Won't Be Reelected In A Recession Equities: Analysts have become overly pessimistic about the earnings outlook for this year, cutting 2019 U.S. EPS growth to 7% (and only 2% YoY in Q1). Our top-down model (based on, admittedly optimistic, U.S. growth assumptions, but also headwinds from a stronger dollar) indicates 12% growth. If analysts are forced to revise up their numbers as better earnings come through, that should be a catalyst for further equity performance (Chart 8). We continue to prefer U.S. over European equities. The steady slowdown in European growth over the past 12 months has not yet bottomed, banks in Europe remain troubled, the earnings picture is less positive, and valuations relative to the U.S. are not especially attractive. We also remain underweight on EM equities: they may produce a positive return in a risk-on environment, but we see them underperforming DM as rising U.S. interest rates and a stronger USD put pressure on EM borrowers with excess foreign-currency debt. Chart 8Analysts Have Overdone Downward Revisions Fixed Income: The recent fall in U.S. Treasury yields was mainly caused by the inflation expectation component, itself very sensitive (if rather illogically so) to the oil price (Chart 9). As the oil price recovers (see below), inflation picks up moderately, and the Fed hikes by more than the market expects, we see the 10-year Treasury yield rising to 3.5% during the course of the year. BCA’s fixed-income strategists recently raised their recommendation on global credit to overweight, given more attractive spreads and the likelihood that the Fed will be on hold for the next six months.2 Their recommendation is for 3-6 months, and the Fed restarting the hiking cycle, say in June, might terminate the positive story. We are following their lead, by raising both high-yield and investment-grade bonds to overweight within the (underweight) fixed-income asset class. That means we are neutral credit in the overall portfolio. We would warn, though, that this is a somewhat short-term call: we still prefer equities as a way to play the continuing risk-on rally. Given the high level of U.S. corporate leverage, and the over-owned nature of the credit market, this is likely to be an asset class that performs very poorly in the next recession (Chart 10). Chart 9Inflation Expectations Should Recover Chart 10Corporate Leverage Is A Concern Currencies: Currencies will continue to be driven by relative monetary policy. With the growth desynchronization between the U.S. and other DMs set to continue (to a degree), we see modest further USD appreciation this year. The Fed (as argued above) will probably hike more than the market expects. But, given slow European growth, the ECB is unlikely to be able to hike in Q4 this year, as it currently is guiding for and the futures market implies (Chart 11). We see the ECB reopening the Targeted Long-Term Repo Facility (TLTRO), which expires soon. Italy and Spain have been big borrowers from this facility, and bank loan growth is likely to slow as it ends (Chart 12). A renewed TLRTO would be seen as a dovish move. Tighter dollar liquidity conditions also point to a stronger USD. U.S. credit growth continues to accelerate (to 12% YoY – Chart 13) in an environment where the monetary policy has tightened: credit growth is outpacing U.S. money supply growth by 7%. Historically this has been negative for global growth (mainly because the deteriorating liquidity is a problem for EM dollar borrowers) and positive for the dollar (Chart 14).3 Chart 11Can ECB Really Hike In 2019? Chart 13...U.S. Loan Growth Accelerating... Chart 14... Which Will Tighten Liquidity Further Commodities: The supply/demand situation for oil should improve over coming months. With Saudi Arabia and Russia committed to cut supply by 1.2 million barrels/day, U.S. shale production growth slowing given the low one-year forward price for WTI, Canada reducing production, and Venezuela on the verge of collapse (which alone could remove 700-800k b/d from the market), our energy strategists see the crude oil balance in deficit over the next four quarters (Chart 15). Given this, they forecast Brent crude rebounding to above $80 a barrel. Other commodity prices are mostly driven by Chinese demand. We see China continuing to slow, until the accumulated effects of its fiscal and mild monetary stimulus start to come through in H2 and stabilize growth. Our analysis suggests that China remains very disciplined about the size and nature of its stimulus: it is not turning on the liquidity taps as it did in early 2016. Bank loan growth has stabilized, but shadow banking activity continues to contract, as the authorities persist with their crackdown and their emphasis on deleveraging (Chart 16). Industrial commodities prices are therefore likely to weaken over the next six months.  Chart 15Oil Balance In Deficit This Year Chart 16China Sticking To Credit Crackdown   Garry Evans, Senior Vice President Global Asset Allocation garry@bcaresearch.com   GAA Asset Allocation Footnotes 1      Please see Geopolitical Strategy Weekly Report, “So Donald Trump Cares About Stocks, Eh?”, dated 9 January 2019, available at gps.bcaresearch.com 2      Please see Global Fixed Income Strategy Weekly Report, “Enough With The Gloom: Upgrade Global Corporates On A Tactical Basis,” dated 15 January 2019, available at gfis.bcaresearch.com 3      For a detailed explanation, please see Foreign Exchange Strategy Weekly Report, “Global Liquidity Trends Support The Dollar, But…,” dated 25 January 2019, available at fes.bcaresearch.com

The hiatus in the Fed’s rates-normalization policy in 1H19 in the wake of its capitulation to financial markets, supports our bullish view on gold prices, as it raises the risk of an inflation overshoot later this year. Per the Fed’s dual mandate, inflation and employment gauges are signaling the need for tighter policy, according to BCA’s proprietary Fed Monitor. The pause in hiking fed funds raises the likelihood the Fed will find itself behind the inflation curve, as the economy enters a late-cycle phase. Gold will outperform other commodities and equities in this phase. We remain long gold as a portfolio hedge. Highlights Energy: The U.S. imposed sanctions on state-owned Petróleos de Venezuela, S.A. (PDVSA), including a ban on the company’s Houston-based Citgo remitting earnings back to the parent company.  This raises the likelihood production and exports will fall sharply as we expect.  Separately, Saudi Energy Minister Khalid al-Falih said the country will reduce output below its recently agreed 10.3mm b/d cap in 1H19, in line with our own balances expectation.1 Base Metals/Bulks: Neutral.  Iron ore prices likely will continue to move higher, following the collapse of a wet-processing dam at Vale’s Córrego do Feijão mine.  The company suffered a similar breach at its Samarco mine in March 2016, which still has not re-opened. Output will fall, if it follows through with additional dam closures. Precious Metals: Neutral.  Gold prices will continue to move higher, as the Fed’s near-term capitulation on its rates-normalization policy raises the odds the U.S. central bank will find itself behind the inflation curve.  (See below.) Ags/Softs: Underweight.  USDA reported soybeans inspected for export to China during the week ended January 24 accounted for close to 37% of the total beans inspected.  This made China the No. 1 importer of American soybeans again. Feature In February 2018, we wrote that “price risk in gold will remain skewed to the upside this year, even as our base case scenario calls for limited gains from here.” In line with this expectation, we suggested remaining long gold as a portfolio diversifier and hedge against mounting equity risks. This turned out to be an accurate call. Despite losing 8.4% between January and September 2018 because of an aggressive Fed, gold rose by 7.6% in 4Q18 amid the rising equity volatility and ended the year down a minor -1.5% compared to -6.2%, -11.2% and -7.1% for the S&P 500, global equities and the CRB commodity index. This reflects the convexity in gold returns and is the reason we favored gold in 2018. Gold returns are not simply a function of the U.S. dollar and real interest rates. As highlighted in our 2019 Key Views report last December, in mature economic cycles, gold’s ability to hedge against equity and inflation risks dominate its price formation, while its correlation with the U.S. Treasury yields diminishes (Chart of the Week).2 Chart of the WeekGold's Correlation With U.S. Rates Declines As The Cycle Matures As the current cycle extends to 2019, the skewness in gold return will prove profitable. The Fed’s retreat on its quarterly rate-hike cycle only adds to our positive view, as it increases the probability the U.S. central bank falls behind the curve. Stay long gold as a portfolio hedge. Fed’s Short-Term Capitulation Strengthens Our View The recent downward revision in the Fed’s rate-hike path reinforces our positive stance on gold prices, as risks of an overshoot in inflation rises. The dichotomy in U.S. vs. rest of the world growth puts the Fed in a difficult position. The current capitulation was mainly driven by tightening financial conditions – chiefly, the rising U.S. dollar, declining stock prices, and widening credit spreads. However, under the Fed’s dual mandate, inflation and employment still are signaling “tightening-required” per BCA Research’s Fed Monitor, a model maintained by our U.S. Bond strategists (Chart 2). Since economic growth cannot remain above-trend indefinitely, short-term productive capacity constraints (i.e. capital and labor factors of production) are already binding and will force the Fed to raise rates later this year as inflation creeps up. Chart 2Growth And Inflation Signal Tighter Money Is Required As it reaffirms its data dependence, the Fed is opening the door to falling behind the inflation curve, given inflation is a lagging indicator of the price pressures that are building up in the economy (Chart 3). As a result, we expect gold’s ability to hedge against inflation will support its price in 2H19. Chart 3Inflationary Pressure Will Rise In 2019 Short-term, a Fed pause also supports gold by readjusting investors’ expectations regarding the U.S. dollar and real interest rates lower. Our bond strategists identified two previous periods where similar conditions led to a false start in the Fed hiking cycle, 1997 and 2015. In both cases, the Fed’s capitulation led to a reversal in gold’s downward price trajectory, as the market perceived the central bank was keeping its short-term policy rate at a level that was inconsistent with the so-called R-star rate or natural rate of interest – i.e., “the real interest rate expected to prevail when the economy is at full strength” (Chart 4).3 Chart 4AGold Price's Trajectory Reversed In 1997... Chart 4B Using a conceptual four-quadrant framework developed by our colleagues at The Bank Credit Analyst to describe the Fed’s behavior, we currently believe the outcome with the highest probability of being realized by the Fed’s capitulation is Policy Mistake 2 (Table 1, lower right quadrant). If we’re right, this raises the odds of an inflation overshoot above the Fed’s 2% target later this year.4 Table 1Four Fed Policy Scenarios This is not a foregone conclusion. However, generally speaking, the higher the inflation uncertainty and the higher the perception the Fed will fall behind the curve, the higher gold is bid up. Recent price action seems to corroborate this. Chart 5 shows that the recent downward revision in the median long-term fed funds rate projection coincides with a rise in gold prices. At present, gold investors are signaling that the fed funds rate is below the neutral rate consistent with R-star. Chart 5Gold Markets Signal Monetary Policy Is Accommodative Gold And The U.S. Economic Cycle Gold prices are difficult to model and predict, given the collection of time-varying, often conflicting, components determining their evolution. Its core determinants change as we move through the economic cycle. In their current late-cycle environment, inflation and equity risks – i.e., fears of a sharp correction – usually gain in importance. In this report, we characterize the market’s late-cycle phase using two metrics: (1) the fed funds rate relative to R-star, (2) the phase of the yield curve cycle.5 We have already discussed (1) in our outlook and found that when the fed funds rate is rising yet still below the estimate of R-star, gold returns are highly skewed to the upside (Chart 6).6 For (2), we compared the yellow metal’s return to other assets returns in different phases of the U.S. Treasury yield curve’s evolution. We define these yield-curve phases as follow: Phase 1: Normal (i.e., positively sloped: 10-year rates are greater than 3-month rates). The 3-month/10-year treasury slope is above 75 bps. Phase 2: On its way to flattening and returning to normal. The 3-month/10-year Treasury slope is between 0 bps and 75 bps. We divide this in two sub-phases: (a) steepening, and (b) flattening. Phase 3: Inverted (i.e., negatively sloped). The 3-month/10-year Treasury slopes is below 0 bps (Chart 7).7 Chart 7Phases Of The Yield Curve Cycle We found that: first, DM and EM equities are the best performers in the group we looked at during Phase 1, when the slope of the yield curve is steep (above 75 bps). Second, there is wide difference between the steepening and flattening sections of Phase 2. EM equities and copper experience the largest rebound once the slope’s curve steepens from below zero. Lastly, gold performs best in the flattening section of Phase 2 and, critically, it outperforms oil, copper, broad commodity indices and equities (Table 2). Table 2Gold Returns Are Positive When The Yield Curve’s Slope Flattens Our U.S. Investment and Bond Strategists believe the Fed’s policy rate will remain in the below-r-star-and-rising range, and in Phase 2 of the yield curve cycle for most of 2019. We agree, and believe our analysis indicates gold prices will increase this year on the back of these factors. Recession Fear And Equity Risks Will Drive Gold For most of 2018, investor sentiment and positioning were primarily determined by the U.S. dollar and real rates. As these variables rose last year, investors’ sentiment and positioning turned overly bearish; this pushed our Gold Composite Indicator in the oversold territory (Chart 8).8 In our view, the other (important) drivers of gold prices were ignored during that period. The end-of-year equity selloff led to a reshuffle of the core determinants of the yellow metal’s price, pushing the equity risk factor higher on the list of variables explaining its price. Chart 8Sentiment Collapsed In 1H18 Chart 9 shows gold and the U.S. equity risk premium disconnected in 2018, until the October equity selloff. In general, these variables are positively linked. When risk aversion is elevated, investors demand higher compensations for holding risky assets, and increase their demand for safe-haven assets. This pushes up both the equity risk premium and gold prices. Chart 9Gold And Equity Risk Premium Correlation Picked Up Gold’s performance in 4Q18 supports our recommendation for holding it as a portfolio diversifier in 2018, and why we continue to do so this year (Chart 10). Separately, our U.S. dollar and rates-only model moved up recently, easing the downward pressure on gold (Chart 11). While we believe these two variables’ marginal impact diminished since 4Q18, they are included in our gold “fair-value” model, which currently indicates it is fairly valued and that its support remains intact. Chart 11Upside Pressures Are Building Bottom Line: The Fed’s near-term capitulation raises the odds the U.S. economy will experience an inflation overshoot. Our fair-value model also is supportive of gold prices. We remain long as a diversification and portfolio hedge. Hugo Bélanger, Senior Analyst Commodity & Energy Strategy HugoB@bcaresearch.com Robert P. Ryan, Senior Vice President Commodity & Energy Strategy rryan@bcaresearch.com Footnotes 1      Please see “Saudis Pledge Deeper Oil Cuts in February Under OPEC+ Deal,” published by bloomberg.com January 29, 2019.  See also “OPEC Starts Cutting Oil Output; Demand Fears Are Overdone” published January 24, 2019, for our latest supply-demand balances and price forecasts.  It is available at ces.bcaresearch.com. 2      Please see BCA Research’s Commodity & Energy Strategy Weekly Report titled “2019 Key Views: Policy-Induced Volatility Will Drive Markets,” published December 13, 2018. It is available at ces.bcaresearch.com. 3      Please see John C. Williams’s remarks delivered to the Economic Club of Minnesota May 15, 2018, entitled “The Future Fortunes of R-Star: Are They Really Rising?”  Williams was president and CEO of the Federal Reserve Bank of San Francisco at the time, and now has the same role at the NY Fed..  We explore this further below.  See also BCA Research’s U.S. Bond Strategy Weekly Report titled “An Oasis Of Prosperity,” published August 21, 2018. It is available at usb.bcaresearch.com. 4      Please see BCA Research’s The Bank Credit Analyst January 2019 Monthly Report published December 21, 2018. It is available at bca.bcaresearch.com. 5      The San Francisco Fed defines R-star as the inflation-adjusted “natural” rate of interest consistent with a fully employed economy, with inflation close to the Fed’s target. R-star is used to guide interest-rate policy consistent with long-term macro goals set by the Fed. Please see “R-star, Uncertainty, and Monetary Policy,” by Kevin J. Lansing, published in the FRBSF Economic Letter May 30, 2017. 6      We presented this analysis in BCA Research’s Commodity & Energy Strategy Weekly Report titled “2019 Key Views: Policy-Induced Volatility Will Drive Markets,” published December 13, 2018. It is available at ces.bcaresearch.com. 7      For a similar analysis applied to different asset classes, please see BCA Research’s U.S. Bond Strategy Weekly Report titled “2019 Key Views: Implication For U.S. Fixed Income,” published December 11, 2018, and The Bank Credit Analyst January 2019 Monthly Report published December 21, 2018. These reports are available at usb.bcaresearch.com and bca.bcaresearch.com. Our approach is slightly different from our colleagues’ methodology. We used a threshold of 75 bps instead of 50 bps in order to increase the sample size of the Phase 2, flattening section. This improves the accuracy of using the average as our main descriptive statistic. Note that the yield curve can remain inverted for some time before a recession occurs, this explains why equity returns are positive in Phase 3 (curve inversion). 8      Our Gold Composite Indicator has three components: (1) Sentiment, (2) Speculative positioning and (3) Technical. It is meant to assess if there is any mismatch between our fundamental analysis and investors’ sentiment and expectations. Investment Views and Themes Recommendations Strategic Recommendations Tactical Trades TRADE RECOMMENDATION PERFORMANCE IN 4Q18 Commodity Prices and Plays Reference Table Summary Of Trades Closed In 2018

Special Report

After rising for thousands of years, human intelligence has begun to decline in developed economies. This can be seen in falling IQ scores and a decline in math and science test scores. Environmental factors appear to account for the bulk of this decline, but no one knows what these factors are. If left unchecked, falling intelligence will severely undermine productivity growth. This could lead to lower equity multiples, larger budget deficits, and ultimately, much higher government bond yields. Technological advances, particularly in the genetic realm, promise to radically raise IQs. In a complete abandonment of its one-child policy, China will combine these controversial technologies with pro-natal measures in order to boost sagging birth rates. The coming Eugenic Wars will be one of the most important economic and geopolitical developments of the 21st century. Part 1: What The Tame Fox Says In 1959, a Soviet scientist named Dmitry Belyaev embarked on an ambitious experiment: to domesticate the silver fox. A geneticist by training, Belyaev wanted to replicate the process by which animals such as cats and dogs came to live side-by-side with humans. It was a risky endeavor. The Soviets had essentially banned the study of Mendelian genetics in favor of the blank slate ideology that is popular in progressive circles today. Belyaev persevered. Working under the guise of studying vulpine physiology, he selected foxes based on only one trait – tamability. Less than 10% of foxes made it to the subsequent generation, with the other 90% being sent off to fur farms. By the fourth generation, the changes were undeniable. Rather than fleeing humans, the foxes sought out their attention with no prompting whatsoever. They even wagged their tails and whined and whimpered like dogs do. The tame foxes also displayed physical changes. Their ears flopped over. Their snouts became shorter and their tails stood upright. "By intense selective breeding, we have compressed into a few decades an ancient process that originally unfolded over thousands of years," wrote Lyudmila Trut, who began as Belyaev’s assistant and took over the project when her boss died in 1985.  Genetically Capitalist? Evolution can broadly proceed in two ways. The first way is through random mutations. This form of evolution, which scientists sometimes refer to as genetic drift, can take thousands of years to yield any discernable changes. The second way is through natural selection, a process that exploits existing variations in genetic traits. As the Russian fox experiment illustrates, evolution driven by selective pressures (either natural or artificial) can occur fairly quickly. Did selective pressures manifest themselves in human evolution in the lead up to the Industrial Revolution? Did humans, in some sense, domesticate themselves? In his book, A Farewell To Alms, economic historian Gregory Clark argued in the affirmative. Clark documented that members of skilled professions in Medieval England had twice as many surviving children as unskilled workers (Chart II-1). Indeed, the fledgling middle class of the time had even more surviving children than the aristocracy, who were often out fighting wars. As a result, the wages of craftsmen declined by a third relative to laborers between 1200 and 1800, implying that the supply of skilled labor was growing more quickly than the demand for skilled workers over this period. Chart II-1Richer Men Had More Surviving Children In Medieval England In subsequent work, Clark and Neil Cummins argued that the spread of bourgeois values across pre-industrial England was more consistent with a model of genetic transmission than a cultural one (see Box II-1 for details). Similar developments occurred in other parts of the world. For example, in China, the gateway into the bureaucracy for a thousand years was the highly competitive imperial exam. Xi Song, Cameron Campbell, and James Lee showed that high-status men had more surviving children during the eighteenth- and nineteenth-centuries (Chart II-2).1 Chart II-2High-Status Men In China Produced More Descendants In The 18th And 19th Centuries The 10,000 Year Explosion Stephen Jay Gould famously said that “There’s been no biological change in humans in 40,000 or 50,000 years. Everything we call culture and civilization we’ve built with the same body and brain.” Gould was wrong. Data from the International HapMap Project show that human evolution accelerated by 100-fold starting around 10,000 years ago (Chart II-3). Chart II-3Human Evolution Has Accelerated 100-Fold Over The Past 10,000 Years In their book The 10,000 Year Explosion: How Civilization Accelerated Human Evolution, Gregory Cochran and the late Henry Harpending explained why evolution sped up so rapidly.2 The advent of agriculture led to a surge in population levels. This, in turn, increased the absolute number of potentially beneficial genetic mutations that could be subject to selection effects. Farming and the rise of city states also completely reshaped the environment in which people lived. Basic biology teaches us that environmental dislocations of this kind tend to generate selective pressures that cause evolution to accelerate. John Hawks, professor of anthropology and genetics at the University of Wisconsin-Madison, put it best: “We are more different genetically from people living 5,000 years ago than they were different from Neanderthals.” Many of the changes to our genomes relate to diet and diseases. The various genetic resistances that people have built up to malaria are all less than 10,000 years old. Mutations to the LCT gene, which confers lactose tolerance into adulthood, occurred independently in three different geographical locations: one in East Asia, one in the Middle East, and one in Africa. The Middle Eastern variant was probably responsible for the rapid enlargement of the Indo-European language group, which now stretches from India to Ireland. The African variant likely facilitated the Bantu expansion, which started near the present-day border of Nigeria and Cameroon, and then spread out across almost all of sub-Saharan Africa. Evolution Of The Human Brain About half of the genes in the human genome regulate some aspect of brain function. Given the rapid acceleration in evolution, it would be rather surprising if our own brains had not been affected. And indeed, there is plenty of evidence that they were. The frontal lobe of the brain has increased in size over the past 10,000 years. This is the part of the brain that regulates such things as language, memory, and long-term planning. Testosterone levels have also declined. That may explain the steady reduction in violent crime rates (Chart II-4). Chart II-4Long-Term Decline In Homicide Rates We know that certain genes that are associated with higher intelligence have been under recent selective pressure. For example, the gene that leads to torsion dystonia – a debilitating movement disorder – appears to have increased in frequency. Why would a gene that causes a known disease become more widespread? The answer is that individuals who have this particular mutation tend to have IQs that are around 10-to 20-points above the population average. Why IQ Matters IQ has a long and contentious history. Yet, despite numerous efforts to jettison the concept, it has endured for one simple reason: It has more predictive power than virtually anything else in the psychological realm. A simple 30-minute IQ test can help predict future educational attainment, job performance, income, health, criminality, and fertility choices (Table II-1 and Chart II-5). IQ even predicts trader performance!3 Table II-1What IQ Predicts (Results From Meta-Analyses)   Chart II-5IQ Tests Don’t Just Measure How Well You Can Do On An IQ Test Like most physiological traits, IQ is highly heritable.4 The genetic contribution to IQ increases from 20% in early childhood to as high as 80% by one’s late teens and remains at that level well into adulthood.5 This makes IQ almost as heritable as height (Chart II-6). Chart II-6The Inheritance You Cannot Renounce Although there is a great deal of variation among individuals, on average, more intelligent people earn higher incomes (Chart II-7). If the same relationship existed in the pre-industrial era, as seems likely, then human intelligence probably increased in a way that facilitated the economic explosion that we associate with the Industrial Revolution. The stunning implication is that the emergence of the modern era was a question of “when, not if.” Chart II-7IQ Tends To Be Positively Correlated With Income And Wealth Part 2: The Flynn Effect By the late-19th century, it had become clear that the rich were no longer having as many children as the poor. This realization, together with the growing popularity of Darwin’s theories, helped galvanize the eugenics movement. Contrary to popular belief, this movement was not a product of the far-right. In fact, the most vocal proponents of eugenics were among the progressive left. John Maynard Keynes, for example, served as the Director of the British Eugenics Society between 1937 and 1944. Yet, a funny thing happened on the road to idiocracy: The concerns of eugenicists did not come to pass. Rather than becoming dimmer, people became smarter. This phenomenon is now known as the Flynn Effect, named after James Flynn, a psychologist who was among the first to document it. Chart II-8 shows the evolution of IQ scores in a sample of countries between 1940 and 1990. The average country recorded IQ gains of three points per decade over this period, a remarkably large increase over such a relatively short period of time. Chart II-8The Flynn Effect: It Was Nice While It Lasted Explaining The Flynn Effect The Flynn Effect must have been entirely driven by environmental factors since genetic factors – namely the tendency of less-educated people to have more children, and to have them at an earlier age – would have reduced average IQs over the past two hundred years. But how could environmental factors have played the dominant role in light of the strong role of genes discussed above? The answer was proposed by geneticist Richard Lewontin in the 1970s. Lewontin suggested imagining a genetically-diverse sack of seed corn randomly distributed between two large identical fields. One field had fertilizer added to it while the other did not. Genetic variation would explain all of the differences in the height of corn stalks within each field, while environmental factors (the addition of fertilizer) would explain all of the difference in the average height of corn stalks between the two fields. This logic explains why genes can account for the bulk of the variation in IQs within any demographic group, while environmental effects may explain most of the variation across groups, as well as why average scores have changed over time. And what environmental effects are these? The truth is that no one really knows. Plenty of theories have been advanced, but so far there is still little consensus on the matter. Bigger, Healthier Brains It has long been known that learning increases the amount of grey matter in the brain. For example, a recent study showed that the hippocampi of London taxi drivers tend to be larger due to the need for drivers to memorize and navigate complex routes.6 The emergence of modern societies likely kicked off a virtuous circle where the need to solve increasingly complex tasks forced people to hone their learning skills, leading to higher IQs and further technological progress. The introduction of universal primary education amplified this virtuous circle. Better health undoubtedly helped as well. Early childhood diseases reduce IQ by diverting the body’s resources away from mental development towards fighting off infections. There is a strong correlation between measured IQ and disease burden across countries (Chart II-9). A number of studies have documented a strong relationship between the timing of malaria eradication in the U.S. and other parts of the world and subsequent observed gains in childhood IQs.7 Chart II-9IQ And Disease Burden Brain size and IQ are positively correlated. Forensic evidence from the U.S. suggests that the average volume of adult human skulls has increased by 7% since the late 1800s, or roughly the size of a tennis ball.8 Part 3: The End Of A 10,000 Year Trend The problem with environmental effects is that they eventually run into diminishing returns. This appears to have happened with the Flynn Effect. In fact, not only does the recent evidence suggest that the Flynn Effect has ended, but the data suggest that IQs are starting to decline. Chart II-10 shows that average math and science test scores fell in the OECD’s Program For International Scholastic Achievement (PISA) between 2009 and 2015, the latest year of the examination. The drop in math and science test scores has been mirrored in falling IQ scores. Flynn observed a decade ago that IQs of British teenagers were slipping.9 Similar results have been documented in France, the Netherlands, Germany, Denmark, and most recently, Norway. Chart II-10Math And Science Skills Are Slipping In The OECD The Norwegian results, published last year, are particularly noteworthy.10 Bernt Bratsberg and Ole Rogeberg examined three-decades worth of data on IQ tests of Norwegian military conscripts. Military duty has been mandatory for almost all men in Norway since 1814, which means that the study’s authors were able to collect comprehensive data on most Norwegian men and their fathers.  Their paper clearly shows that IQ peaked with the generation born in the mid-1970s and declined by about five points, or one-third of a standard deviation, for the one born in 1990 (Chart II-11). For the first time in recorded history, Norwegian kids today are not scoring as well as their parents. Chart II-11Norwegian IQs Have Been Falling For Those Born Since The Mid-1970s A Mystery What caused the sudden reversal of the Flynn Effect in Norway and most other developed economies? Nobody knows. We can, however, offer three possible theories: New Technologies For much of human history, rising intelligence and technological innovation were complementary processes, meaning that the smartest people were the ones who could best exploit the new technologies that were coming their way. Moreover, as noted above, even those who were less gifted benefited from the mental stimulation that a technologically advanced society provided. It remains to be seen how future technological advances such as generalized AI will affect human intelligence, but recent technological advances seem to have had a dumbing down effect.11 For example, the GPS has obviated the need for people to navigate unfamiliar locations, thus blunting the development of their visuospatial skills. Modern word processors have made spelling skills less important. Having all the information in the world just a click away is a wonderful thing, but it has reduced the need for our brains to retain and codify what we learn. Meanwhile, the constant bombardment of information to which we are subject has made it difficult to concentrate on anything for long. How many youth today can read a report of this length without checking their Facebook feed multiple times? My guess is not many. Diminishing Returns To Education The ability to take young bright minds, who would have otherwise spent their lives doing menial labor, and provide them with an education was probably the greatest tailwind to growth that the 20th century enjoyed. There is undoubtedly still scope to continue this process, but the low-hanging fruits have been picked. Educational attainment has slowed dramatically in most of the world (Chart II-12). Economist James Heckman estimates that U.S. high-school graduation rates, properly measured, peaked over 40 years ago.12 Chart II-12The Contribution To Economic Growth From Rising Human Capital Is Falling Despite billions of dollars spent, efforts to improve school performance have generally fallen flat. A recent high-level report by the U.S. Department of Education concluded that “The panel did not find any empirical studies that reached the rigor necessary to determine that specific turnaround practices produce significantly better academic outcomes.”13 This gets to a point that most parents already know, which is that when people talk about “bad schools," they are really talking about “bad students.” Deteriorating Health Better health probably contributed to the Flynn Effect. But is it possible to have too much of a good thing? More calories are welcome when people are starving, but today’s calorie-rich, nutrient-poor diets have led to a surge in obesity rates. A clean environment reduces the spread of germs, but it also makes children hypersensitive to foreign substances. Following German reunification, researchers observed that allergies were much more common among West German children than their Eastern peers, presumably because of the West’s more salubrious environment.14 All sorts of weird and concerning physiological changes are occurring. Sperm counts have fallen by nearly 60% since the early 1970s.15 Testosterone levels in young men are dropping. Among girls, the age of first menarche has declined by two years over the past century.16 Are chemical agents in the environment responsible? If they are, what impact are they having on cognitive development? Nobody knows. Reported mental illness is also on the rise. The share of U.S. teenagers with a reported major depressive episode over the prior year surged by over 60% between 2010 and 2017 (Chart II-13). The fraction of young adults that made suicide plans nearly doubled.17 More than 20% of U.S. women over the age of 40 are on antidepressants.18 Five percent of U.S. children are receiving ADHD medication.19 Chart II-13A Disturbing Deterioration In Mental Health Implications For Economic Growth And Asset Markets So far, the reversal of the Flynn Effect has been largely confined to the developed economies. Test scores are still rising in the developing world, albeit from fairly low levels. For example, two recent studies have documented significant IQ gains in Kenya and Brazil.20 In the poorest countries, opportunities for improving health abound. Even small steps such as fortifying salt with iodine (which costs about five cents per person per year) have been shown to boost IQ by nearly one standard deviation.21 Measures to reduce inbreeding are also likely to boost IQ scores.22 Yet, we should not underestimate the importance of falling cognitive skills in developed economies. Chart II-14 shows that there is a clear positive correlation between student score on math and science and per capita incomes. Chart II-14Positive Correlation Between Student Scores On Math And Science And Per Capita Income Most technological innovation still takes place in developed economies. There is an extremely tight relationship between visuospatial IQ and the likelihood of becoming an inventor (Chart II-15). Since IQ is distributed along a bell curve, a 0.1 standard deviation drop in IQs across the entire distribution will result in an 8% decline in the share of people with IQs over 100, a 14% decline in those with IQs over 115, and a 21% decline in those with an IQ over 130 (by convention, each standard deviation on an IQ test is worth 15 points). Chart II-15Having A Compass In Your Head Makes You A Better Inventor Falling IQs could result in slower productivity growth, which could further strain fiscal balances. Lower IQs are also associated with decreased future orientation.23 People who live for the moment tend to save less. A decline in savings would push up real rates, leading to less capital accumulation. History suggests that a deceleration in productivity growth and higher real rates will put downward pressure on equity multiples (Chart II-16). Chart II-16Equity Multiples Tend To Fall When Real Rates Rise And Productivity Growth Declines Part 4: Generation E For 200 years, the environmentally-driven Flynn Effect disguised the underlying genetically-driven decline in IQs that began not long after the dawn of the Industrial Revolution. Flynn has acknowledged this himself, noting at the 2017 International Society For Intelligence Research Conference that “I have no doubt that there has been some deterioration of genetic quality for intelligence since the late Victorian times.”24 Now that the Flynn Effect has reversed, both genes and the environment are working together to reduce cognitive abilities in developed economies. This means that the most important trend in the world – a trend that allowed the human population to increase during the Malthusian era and later allowed output-per-worker to soar following the Industrial Revolution – has broken down. Yet, there may be another twist in the story – one that began just a few months ago: the first members of Generation E were born. E Is For Edited ... Or Eugenics Lulu and Nana will be like most other children, but with one key difference: They will be the first humans ever to have their genomes edited through a procedure know as CRISPR-Cas9. Rogue Chinese scientist He Jiankui deactivated their CCR5 gene, which the HIV virus uses as a gateway into the body. His actions were rightfully condemned around the world for endangering the twins’ health by using a procedure that has not yet been fully vetted in animal studies, let alone in human trials (Lulu and Nana’s father is HIV+ but it is debatable whether the children were at an elevated risk of infection). He Jiankui remains under house arrest at the university where he worked. But whatever his fate, the dam has been broken. For better or for worse, the era of personal eugenics has arrived. The Return Of The Silver Fox It is easier to delete a gene than to add one. It is even more difficult to swap out a large number of genes in a way that achieves a predictable outcome. Thus, the successful manipulation of highly polygenic traits such as intelligence — traits that are linked to hundreds of different genes – may still be decades away.25 Predicting a trait is much simpler than modifying it, however. The cost of sequencing a human genome has fallen by more than 99% since 2001 (Chart II-17). Start-up company Genomic Prediction has already developed a test for fertilized embryos for IVF users that predicts height within a few centimetres and IQ with a correlation of 0.3-to-0.4, roughly as accurate as standardized tests such as the SAT or ACT.26 Other companies are following suit.27 Chart II-17Sequencing Costs Have Plummeted Some will recoil in horror at the prospect of selecting prospective children in this manner. They will argue that such technologies, beyond being simply immoral, will widen social inequality between those who can afford them and those who cannot. Others will counter that screening embryos for certain traits is not that dissimilar to what people already do with prospective romantic partners. They will also point out that mass usage of these technologies will drive down prices to the point that even poor people will be able to access them, thus giving low IQ parents the chance to have high IQ kids. They might also note that such technologies may be the only way to reverse the ongoing accumulation of deleterious mutations within the human germline that has been the unintended by-product of the proliferation of life-saving medicines.28 We will not wade into this thorny debate, other than to note that there will be huge incentives for people to avail themselves of these technologies. The Coming Eugenic Wars And not just individuals either – governments too. While the initial impact of eugenic technologies will be small, the effects will compound over time. Carl Shulman and Nick Bostrom estimate that genetic screening could boost average IQs by up to 65 points in five generations (Table II-2). Table II-2A Poisoned Chalice? Genetic Screening Can Raise IQ China has been investing heavily in genetic technologies. As Geoffrey Miller has argued, China’s infatuation with eugenics spans into the modern day.29 Like most other countries, fertility in China is negatively correlated with IQ. Mingrui Wang, John Fuerst, and Jianjun Ren estimate that China is currently losing nearly one-third of a point in generalized intelligence per decade, with the loss having accelerated rapidly between the 1960s and mid-1980s.30 The decline in the genetic component of Chinese IQs is coming at a time when the population itself is about to shrink. According to the UN’s baseline forecast, China will lose 450 million working-age people by the end of the century (Chart II-18). Meanwhile, the country is saddled with debt, the result of an economic model that has, for decades, recycled copious household savings into debt-financed fixed-investment spending in an effort to shore up domestic demand. Chart II-18China’s Labor Pool Is Shrinking The authorities may be tempted to tackle all three problems simultaneously by adopting generous pro-natal measures – call it the “at least one-child policy”– which increasingly harnesses emerging eugenic technologies. The resulting baby boom would strengthen domestic demand, thus making the economy less dependent on exports, while ensuring China’s long-term geopolitical viability. The Eugenic Wars are coming, and they will be unlike anything the world has seen before. BOX II-1 The Diffusion Of Bourgeois Values: Culture Or Genes? Higher-income people had more surviving children in the centuries leading up to the Industrial Revolution. Real per capita income was broadly stable during this period. This implies that there must have been downward social mobility, with sons, on average, being less wealthy than their fathers. This downward mobility, in turn, spread the characteristics of higher-income people across the broad swathe of society. What were these characteristics? Cultural values that emphasized thrift, diligence, and literacy were undoubtedly part of what was passed on to future generations. But surprisingly, it also appears that genetic transmission played an important, and perhaps pivotal, role. Models of genetic transmission make very concrete predictions about the correlations in economic status that one would expect to see among relatives. Biological brothers share 50% of their genes, as do fathers and sons. Likewise, first cousins share 25% of their genes, the same as grandfathers and sons. These facts yield two testable predictions: The first is that the correlation coefficient on status measures such as wealth, occupation, and education should be the same for relatives that share the same fraction of genes such as sibling pairs and father-son pairs. Box Chart II-1 shows that this is borne out by the data. The second prediction is that the correlation between status and genetic distance should follow a linear trend so that, for example, the correlation in wealth among brothers is twice that of first cousins and four times that of second cousins. Box Chart II-2 shows that this is also borne out by the data. BOX CHART II-1Comparative Father-Son And Brother Correlations By Status   BOX CHART II-2Wealth Correlations By Genetic Distance Other evidence supports the importance of genes in the transmission of status across generations. The correlation in measures such as wealth, education, and occupation is much higher among identical twins than fraternal twins. Adopted children turn out to be more similar to their biological parents on these measures when they reach adulthood than their adopted parents, even when the children have never met their biological parents. The parent-child correlation also remains the same regardless of family size, suggesting that spreading the same resources over more children may not harm life outcomes to any discernible degree, at least on the measures listed above. Peter Berezin Chief Global Strategist Global Investment Strategy   Footnotes 4       Xi Song, Cameron Campbell, and James Lee, "Descent Line Growth and Extinction From A Multigenerational Perspective, Extended Abstract," American Sociological Review 80:3, (April 21, 2015): 574-602. 5       Gregory Cochran and Henry Harpending, "The 10,000 Year Explosion: How Civilization Accelerated Human Evolution," Basic Books, (2009). 6       Mark Grinblatt, Matti Keloharju, and Juhani T. Linnainmaa, “IQ, Trading Behavior, and Performance,” Journal of Financial Economics, 104:2, (May 2012): 339-362. 7       Thomas Bouchard, "Genetic Influence On Human Psychological Traits - A Survey," Current Directions in Psychological Science 13:4, (August 2004): 148-151. 8      The tendency for the genetic contribution to IQ to increase until early adulthood and then to remain at high levels until old age is known as the Wilson Effect. There is no consensus on what causes it, but it probably reflects a number of factors: 1) It may take some children longer than normal to reach full intellectual maturity. Testing their IQs at a young age will result in scores that are lower than those expected based on their parents’ IQs. The opposite is true for children whose IQs increase relatively quickly in young age, but possibly top out earlier; 2) Environmental effects are probably more important in young age when a child’s brain is still quite malleable; 3) Self-reinforcing gene-environment interactions tend to increase with age. Children do not have much control over their environment, but as they get older, they will seek out activities that are more in keeping with their genetic predispositions. For example, a studious child may pursue a career that reinforces their love of learning. 9       "Cache Cab: Taxi Drivers' Brains Grow to Navigate London's Streets," Scientific American, (December 2011). 10       Atheendar Venkataramani, "Early Life Exposure to Malaria and Cognition in Adulthood: Evidence from Mexico," Journal of Health Economics 31:5, (July 2012): 767-780; Hoyt Bleakley, "Health, Human Capital and Development," Annual Review of Economics 2, (March 2010): 283-310; Hoyt Bleakley, "Malaria Eradication in the Americas: A Retrospective Analysis of Childhood Exposure," American Economic Journal: Applied Economics 2, (April 2010): 1-45. 11       "Anthropologists Find American Heads Are Getting Larger," ScienceDaily, (May 2012). 12       "British Teenagers Have Lower IQs Than Their Counterparts Did 30 Years Ago," The Telegraph, (February 2009). 13     Bernt Bratsberg and Ole Rogeberg, "Flynn Effect And Its Reversal Are Both Environmentally Caused," Proceedings of the National Academy of Sciences 115:26, (June 2018): 6674-6678. 14     On the face of it, artificial intelligence would appear to be a substitute for human intelligence. Many applications of AI would undoubtedly have this feature, especially those that allow computers to perform complex mental tasks that humans now must do. However, there are several ways that AI may eventually come to complement human intelligence. First, and most obviously, AI could be used to augment human capabilities either directly by hardwiring it into our brains, or indirectly through the development of drugs or genetic techniques which improve cognition. Second, looking further out, the benefits of highly intelligent AI systems would be limited if humans did not possess the requisite intelligence to understand certain concepts that are currently beyond our mental reach. No matter how well intentioned, trying to explain string theory to a mouse is not going to succeed. There are probably a multitude of ideas that AI could reveal that we simply cannot comprehend at current levels of human intelligence. 15     James Heckman and Paul La Fontaine, "The American High School Graduation Rate: Trends and Levels," The Review of Economics and Statistics 92:2, (May 2010): 244–262. 16     "Turning Around Chronically Low-Performing Schools," The Institute of Education Sciences (IES), (May 2008). 17     E. von Mutius, F.D. Martinez, C. Fritzsch, T. Nicolai, G. Roell, and H. H. Thiemann, "Prevalence Of Asthma And Atopy In Two Areas Of West Germany And East Germany," American Journal of Respiratory and Critical Care Medicine 149:2, (February 1994): 358-64. 18     "Sperm Counts In The West Plunge By 60% In 40 Years As ‘Modern Life’ Damages Men’s Health," Independent, (July 2017). 19     Kaspar Sørensen, Annette Mouritsen, Lise Aksglaede, Casper P. Hagen, Signe Sloth Mogensen, and Anders Juul, "Recent Secular Trends in Pubertal Timing: Implications for Evaluation and Diagnosis of Precocious Puberty," Hormone Research in Paediatrics 77:3, (May 2012): 137-145. 20     “Results from the 2017 National Survey On Drug Use And Health: Detailed Tables,” Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality, Rockville (Maryland), (September, 2018). 21     Laura A. Pratt, Debra J. Brody, and Qiuping Gu, "Antidepressant Use Among Persons Aged 12 and Over: United States, 2011–2014," NCHS Data Brief No. 283, Centers for Disease Control and Prevention, (August 2017). 22     Some, but not all, of the increase in reported rates of mental illness may be due to more aggressive diagnosis by health practitioners. For example, a recent study revealed that children born in August were 30% more likely to receive an ADHD diagnosis than those born in September, simply because they were less mature compared to other kids in the first few years of elementary school. See: Timothy J. Layton, Michael L. Barnett, Tanner R. Hicks, and Anupam B. Jena, "Attention Deficit-Hyperactivity Disorder and Month of School Enrollment," New England Journal of Medicine 379:22, (November 2018): 2122-2130. 23     Tamara C. Daley, Shannon E. Whaley, Marian D. Sigman, Michael P. Espinosa, and Charlotte Neumann, "IQ On The Rise: The Flynn Effect In Rural Kenyan Children," Psychological Science 14:3, (June 2003): 215-9; Jakob Pietschnig and Martin Voracek, "One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909-2013)," Perspectives on Psychological Science 10:3, (May 2015): 282-306. 24     N. Bleichrodt and M. P. Born, “Meta-Analysis of Research on Iodine and Its Relationship to Cognitive Development,” In: ed. J. B. Stanbury, "The Damaged Brain of Iodine Deficiency," Cognizant Communication Corporation, New York, (1994): 195-200; "Iodine status worldwide: WHO Global Database on Iodine Deficiency," World Health Organization, Geneva, (2004). 25     Mohd Fareed and Mohammad Afzal, "Estimating the Inbreeding Depression on Cognitive Behavior: A Population Based Study of Child Cohort," PLOS ONE 9:12, (October 2015): e109585. 26     H. de Wit, J. D. Flory, A. Acheson, M. McCloskey, and S. B. Manuck, "IQ And Nonplanning Impulsivity Are Independently Associated With Delay Discounting In Middle-Aged Adults," Personality and Individual Differences 42:1, (January 2007): 111-121; W. Mischel and R. Metzner, "Preference For Delayed Reward As A Function Of Age, Intelligence, And Length Of Delay Interval," Journal of Abnormal and Social Psychology 64:6, (July 1962): 425-31. 27     James Flynn, “IQ decline and Piaget: Does the rot start at the top?” Lifetime Achievement Award Address, 18th Annual meeting of ISIR, (July 2017). 28     For a good discussion of these issues, please see Richard J. Haier, “The Neuroscience of Intelligence,” Cambridge Fundamentals of Neuroscience in Psychology, (December 2016). 29     "The Future of In-Vitro Fertilization and Gene Editing," Psychology Today, (December 2018). 30     "DNA Tests For IQ Are Coming, But It Might Not Be Smart To Take One," MIT Technology Review, (April 2018). 31     Michael Lynch, "Rate, Molecular Spectrum, And Consequences Of Human Mutation," Proceedings of the National Academy of Sciences 107:3, (January 2010): 961-968. 32     Geoffrey Miller, "What *Should* We Be Worried About?" Edge, (2013). 33     Mingrui Wang, John Fuerst, and Jianjun Ren, "Evidence Of Dysgenic Fertility In China," Intelligence 57, (April 2016): 15-24.