Debt Trends

Dear Client, In lieu of next week’s regular report, we will be bringing you a Special Report featuring a no-holds-barred debate over the economic and financial market outlook among three of BCA’s more bullish strategists (Doug Peta, Rob Robis, and yours truly) and three of the more bearish ones (Anastasios Avgeriou, Arthur Budaghyan, and Dhaval Joshi). Best regards, Peter Berezin, Chief Global Strategist Highlights Slowdowns are much more likely to turn into recessions when significant economic and financial imbalances are present. The U.S. does not currently suffer from any of the three major imbalances that have historically heralded recessions – rapid private-sector debt growth; excessive spending in cyclical sectors such as housing, consumer durables, and business capex; or accelerating inflation. Imbalances are larger abroad, but not to the extent that they will trigger a global recession. The combination of ongoing Chinese stimulus and the lagged effect from lower bond yields will lift global growth during the coming months. The inventory cycle, which is likely to subtract at least one full percentage point from U.S. growth in Q2, will also turn from being a headwind to a tailwind. Stay overweight global equities relative to government bonds over the next 12 months. A rebound in global growth will push down the U.S. dollar later this year, creating an opportunity to increase exposure to European and EM equities. Feature Global Growth At A Critical Juncture The global economy has clearly slowed since early 2018 (Chart 1). So far, much of the weakness has been confined to the manufacturing sector. However, the service sector has softened as well (Chart 2). Chart 1The Global Economy Has Slowed... Chart 2...Mostly Due To Another Manufacturing Downturn     Regionally, the U.S. has held up somewhat better than most other economies. Nevertheless, the ISM manufacturing and nonmanufacturing indices have both declined, with the former now flirting with the 50 line. All recessions begin as slowdowns but not all slowdowns end in recessions. As we discuss below, slowdowns are much more likely to morph into recessions when financial and economic imbalances are elevated. We confine our empirical analysis to the U.S., but discuss the global context later in the report. Three Key Recessionary Imbalances Three imbalances, in particular, have often been present at the outset of U.S. recessions (Chart 3): Chart 3What Makes A Slowdown Degenerate Into A Recession: Imbalances Rapid private-sector debt growth: Rising debt lifts aggregate demand.1 Fast debt growth is also often associated with bad lending decisions, which makes economies more vulnerable to adverse shocks. An unsustainably high level of cyclical spending: Cyclical spending includes business and residential investment, as well as spending on consumer durable goods. If spending on these categories is elevated, there is more scope for it to decline when the economy turns down. High and rising inflation. When inflation rises above the Fed’s comfort zone, the central bank normally needs to raise rates into restrictive territory.  Fast debt growth is also often associated with bad lending decisions, which makes economies more vulnerable to adverse shocks. Table 1 shows every episode since 1960 when the U.S. economy has slowed significantly. To keep things simple, we define a slowdown as a 10-point drop in the ISM manufacturing index from its recent high. Table 1Episodes Of Significant Economic Slowdown Of the 15 slowdowns that we examined, seven culminated in recessions. An average of 2.1 of the three imbalances listed above were visible prior to recessions. However, an average of only 0.9 imbalances were present when a recession failed to materialize. This supports our claim that slowdowns are more likely to turn into recessions when significant imbalances are present. The good news for the U.S. is that it currently does not register any of three imbalances that have typically preceded recessions. Equities reacted very differently in the two cases. When a recession did occur following the start of a slowdown, the S&P 500 declined by an average of 3.6% over the subsequent 12 months. When the slowdown failed to turn into a recession, the S&P rose by an average of 18.3%. In the latter case, the recovery in stocks usually coincided with a swift rebound in the ISM index. The U.S. Is Currently 0 For 3 On The Imbalance Front The good news for the U.S. is that it currently does not register any of three imbalances that have typically preceded recessions. Chart 4Reasons Not To Panic About U.S. Corporate Debt (I) Private-Sector Debt While U.S. private nonfinancial debt has edged up slightly as a share of GDP since 2015, it remains well below its 2008 peak. In fact, the current business expansion is the only one in the post-war era where private-sector debt has failed to rise above its previous cycle high. A recent Bank of England study examined 130 recessions across 26 countries. It found private debt growth matters much more for recession risk than the level of debt.2 Granted, the composition of debt also matters: While household debt in the U.S. has fallen over the past decade, corporate debt has risen. As a share of GDP, corporate debt is now at the highest level in the post-war era. That said, despite its recent ascent, the ratio of corporate debt-to-GDP is less than two percentage points higher than it was in 2008. One drawback of comparing debt to GDP is that the former is a stock variable while the latter is a flow variable. A more sensible “apples-to-apples” approach is to look at corporate debt in relation to assets rather than GDP. If one does that, one sees that the ratio of U.S. corporate debt-to-assets is below its post-1980 average and only slightly above its post-1950 average. The interest coverage ratio, which compares the profits that companies earn for every dollar of interest that they pay, is above its historic norm (Chart 4). Corporate sector free cash flow – the difference between profits and spending on such things as labor and capital goods – remains in surplus. Every recession during the past 50 years has begun when the free cash flow balance was in deficit (Chart 5). In contrast to mortgages, which are generally held by leveraged institutions such as banks, most corporate debt is held by entities such as insurance companies, pension funds, mutual funds, and ETFs. Banks hold only 18% of corporate debt, down from 40% in 1980 (Chart 6). Thus, while high corporate debt levels could exacerbate the next recession, they are unlikely to engender it.  Chart 5Reasons Not To Panic About U.S. Corporate Debt (II) Chart 6Banks Have Reduced Their Exposure To The Corporate Sector   Cyclical Spending Unlike a restaurant meal or a vacation, a house, office tower, factory, or automobile will usually retain some value for a while after it is purchased. If spending on cyclical items rises to a high level for an extended period of time, a glut will form, requiring a period of lower production. By contrast, if spending on these items is subdued for a long time, pent-up demand will accumulate, requiring a period of higher production.  Recessions can result from either economic overheating or financial market overheating. As a share of GDP, cyclical spending is still far below the peaks observed during past expansions. Just as importantly, today’s low level of cyclical spending follows ten years of even lower spending. As a result, the average age of the U.S. capital stock has increased across almost all categories since 2008 (Chart 7). Most notably, the average age of U.S. homes has risen by nearly five years since 2006, the sharpest increase since the Great Depression. Despite the rebound in residential investment from its recessionary lows, the current level of homebuilding still falls short of what is necessary to keep up with household formation. As a consequence, the vacancy rate has fallen to multi-decade lows (Chart 8). Chart 7The Capital Stock Is Aging Chart 8There Is No Glut Of U.S. Homes   Inflation Recessions can result from either economic overheating or financial market overheating. Economic overheating was the dominant driver of recessions between the late 1960s to early 1980s. Rising inflation preceded the recessions of 1969-70, 1973-75, as well as the back-to-back recessions in 1980-82. Chart 9The 1990 Recession: A Bit Of Everything Overheating also contributed to the 1990 recession. After peaking in 1982, the unemployment rate fell to 5% in 1989, about one percent below its equilibrium level at the time. Core inflation began to accelerate, reaching 5.5% by August 1990. The Fed initially responded to the overheating economy by hiking interest rates. The fed funds rate rose from 6.6% in March 1988 to a high of 9.8% by May 1989. By the summer of 1990, the economy had already slowed significantly. Commercial real estate, still reeling from the effects of the Savings and Loan crisis, weakened sharply. Defense outlays continued to contract following the collapse of the Soviet Union. The final straw was Saddam Hussein’s invasion of Kuwait, which caused oil prices to surge and consumer confidence to plunge (Chart 9). In contrast to earlier downturns, the last two recessions were more the byproduct of financial excesses: The 2007-09 recession stemmed from the housing crash and the financial crisis it generated; the 2001 recession followed the dotcom bust, which precipitated a steep decline in capital spending. What will the next U.S. recession look like? Given the absence of major financial imbalances, the odds are high that the next recession will be a “retro recession,” featuring classic economic overheating. The fact that the Fed has adopted a risk-based approach to monetary policy, which puts great weight on avoiding a deflationary outcome, only raises the likelihood that inflation will eventually move higher. The good news is that this is unlikely to happen anytime soon. While wage growth has picked up, productivity growth has risen even more. As a result, unit labor costs – the ratio of wages-to-productivity – have actually decelerated over the past 18 months. Unit labor cost inflation tends to lead core inflation by up to one year (Chart 10). Given the absence of major financial imbalances, the odds are high that the next recession will be a “retro recession,” featuring classic economic overheating. As we discussed in our latest Strategy Outlook, the Fed will probably not bring rates into restrictive territory until early 2022. This gives the economy plenty of breathing space.3 The Global Dimension The discussion above has focused on the United States. To some extent, this is unavoidable. Not only is the U.S. still the world’s largest economy, but it remains at the heart of the global financial system. U.S. equities account for over half of global stock market capitalization, up from a third in the early 1990s (Chart 11). The dollar continues to be the preeminent reserve currency. As a result, U.S. financial markets drive overseas markets much more than the other way around. Chart 10No Imminent Threat Of A Wage-Price Inflationary Spiral Chart 11The U.S. Stock Market Capitalization Is More Than Half Of Global   This does not mean that the rest of the world is irrelevant. The global supply chain now dominates international trade. More than half of all cross-border trade is in intermediate goods (Chart 12). Irrespective of the financial and economic imbalances discussed above, a full-blown trade war would upend the global economy, sending the U.S. and the rest of the world into recession. President Trump’s re-election prospects would plummet if U.S. unemployment rose and the stock market plunged. This is the main reason for thinking that the trade talks will ultimately produce some sort of détente. Nevertheless, a severe deterioration of trade relations remains the biggest risk to our bullish view on risk assets. The fact that financial and economic imbalances are generally larger overseas means that the rest of the world is more vulnerable to adverse shocks. Unlike in the United States, private debt has risen sharply as a share of GDP in several key economies over the past decade (Chart 13). Government debt is also a problem in countries such as Italy that do not have central banks which can function as reliable lenders of last resort. Chart 14Economies With Frothy Housing Markets Risk Having Deeper Downturns Cyclical spending is fairly elevated in a number of countries. Notably, residential investment stands at near record highs as a share of GDP in Canada, Australia, and New Zealand (Chart 14). Home prices are also quite frothy there. When the global economy falls into recession in two-to-three years, these economies will take it on the chin. Investment Conclusions Notwithstanding the risks noted above, we continue to maintain a bullish outlook on global equities and spread product over the next 12 months. To paraphrase Wayne Gretzky, one should invest on the basis of where the economic data is going, not where it is.4 While global growth remains anemic today, the combination of Chinese stimulus and the lagged effect from lower bond yields will boost activity during the coming months. The inventory cycle, which is likely to subtract at least one full percentage point from U.S. growth in Q2, will also turn from being a headwind to a tailwind. Global equities are not super cheap, but they are not particularly expensive either. The MSCI All-Country World Index trades at 15.3-times forward earnings. Given the ultra-low level of global bond yields, this generates an equity risk premium (ERP) that is well above its historical average (Chart 15). From an asset allocation perspective, one should favor stocks over bonds when the ERP is high. Chart 15AEquity Risk Premia Remain Elevated (I) Chart 15BEquity Risk Premia Remain Elevated (II)   The ERP is especially elevated outside the United States. This is partly because non-U.S. stocks trade at a meager 13.3-times forward earnings, but it also reflects the fact that bond yields are lower overseas. The fact that financial and economic imbalances are generally larger overseas means that the rest of the world is more vulnerable to adverse shocks. As global growth accelerates, the dollar will start to weaken (Chart 16). EM and European equities usually outperform the global benchmark in that environment (Chart 17). We expect to upgrade stocks in these regions later this summer. Chart 16The Dollar Is A Countercyclical Currency Chart 17EM And Euro Area Equities Outperform When Global Growth Improves   Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com   Footnotes 1      Recall that GDP is a flow variable (how much production takes place every period), whereas credit is a stock variable (how much debt there is outstanding). By definition, a flow is a change in a stock. Thus, credit growth affects GDP and the change in credit growth affects GDP growth. 2      Jonathan Bridges, Chris Jackson, and Daisy McGregor, "Down in the slumps: the role of credit in five decades of recessions," Bank Of England Staff Working Paper No. 659, (April 2017). 3      Please see Global Investment Strategy Strategy Outlook, "Third Quarter 2019 Strategy Outlook: The Long Hurrah," dated June 28, 2019. 4      According to Wayne Gretzky, his father, Walter, once advised him to “skate to where the puck is going, not to where it is.”   Strategy & Market Trends MacroQuant Model And Current Subjective Scores Tactical Trades Strategic Recommendations Closed Trades

Special Report

Dear Client, Credit in China has expanded at an exponential pace, with the country’s debt-to-GDP ratio climbing from 143% to more than 250% over the last decade. The speed and scale of China’s debt surge dwarfs Japan and the U.S.’ respective credit binges in the 1980’s and 2000’s, each of which ultimately led to financial market meltdowns. Why should China’s experience be any different? Given that China has pursued a different economic model whereby the banking sector is largely state-sponsored and the currency is tightly managed by the central bank, the answer to this pressing question for global markets is the subject of spirited debate at BCA and within the investment community at large. Clients are already aware that my colleagues, Peter Berezin and Arthur Budaghyan, disagree on the macro and market ramifications of China’s decade-long credit boom. The aim of this report is to provide visibility on the root sources of the view divergence, not to reconcile the gaps. We hope these insights will help shape your own conviction about this important topic. Caroline Miller Global Strategy Feature   Caroline: Arthur, your cautious outlook towards emerging markets in general and China’s prospects in particular stems from your belief that China’s economy is dangerously addicted to credit as a growth driver. Please explain why you dismiss the more sanguine view that China’s elevated debt burden is a function of an equally unusually high household savings rate. Arthur: It is simple: When people use the word “savings,” they typically and intuitively refer to bank deposits or securities investments; but this is incorrect. Chart 1 (Arthur)No Empirical Evidence That Deposits = 'Savings' Money supply/deposits in the banking system have no relationship with the savings rate of a nation in general or households in particular (Chart 1). When households save, they do not change the amount of money supply and deposits. Hence, households’ decision to save neither alters liquidity in the banking system nor helps banks to originate loans. In fact, banks do not intermediate deposits into loans or savings into credit.1 The terms “savings” in economics does not denote an increase in the stock of money and deposits. The term “savings” in economics means the amount of goods and services produced but not consumed. When an economy produces a steel bar, it is registered as national “savings.” We cannot consume (say, eat or expense) a steel bar. Therefore, once a steel bar or any equipment is produced, economic statistics will count it as “savings.” Besides, the sole utilization of a steel bar is in capital goods and construction, and hence, it cannot be consumed. Once a steel bar is produced, both national savings and investment will rise. That is how the “savings” = “investment” identity is derived. Chart 2 (Arthur)Chinese Households Are More Leveraged Than U.S. Ones It would avoid confusion and help everyone if economists were to call it “excess production” not “excess savings.” Banks do not need “excess production” – i.e., national “savings” – to create loans. Critically, the enormous amount of bank deposits in China is not due to household “savings” but is originated by banks “out of thin air.” In fact, Chinese households are now more leveraged than U.S. ones (Chart 2). The surge of credit and money supply in China during the past 10 years has been due to animal spirits running wild among lenders and borrowers on the mainland, not its households’ “savings.” In short, the root of China’s credit bubble is not any different from Japan’s (in the 1980s), or the U.S.’ (in the 2000s) and so on. Peter: Yes, banks can create deposits “out of thin air,” as Arthur says. However, people must be willing to hold those deposits. The amount of deposits that households and businesses wish to hold reflects many things, including the interest rate paid on deposits and the overall wealth of the society. The interest rate is a function of savings. The more people save, the lower interest rates will be. And the lower interest rates are, the more demand for credit there will be (Chart 3). It’s like asking what determines how many apples are consumed. Is it how many apple trees farmers want to plant or how many apples people want to eat? The answer is both. Prices adjust so that supply equals demand. How about national wealth? To a large extent, wealth represents the accumulation of tangible capital – factories, plant and machinery, homes and office buildings: the sort of stuff that banks can use as collateral for lending. And what determines how much tangible capital a country possesses? The answer is past savings, of the exact sort Arthur is referring to: the excess of production over consumption. So this form of “economic” savings also plays an important indirect role in determining the level of bank deposits. Chart 4 (Peter)China: From Exporting Savings To Investing Domestically And Building Up Debt I think the main problem with Arthur’s argument is that he is observing an accounting identity, which is that total bank assets (mostly loans) must equal liabilities (mostly deposits and capital) in equilibrium, without fully appreciating the economic forces – savings being one of them – that produce this equilibrium. In any case, the whole question of whether deposits create savings or savings create deposits misses the point. China’s fundamental problem is that it does not consume enough of what it produces. In the days when China had a massive current account surplus, it could export its excess savings abroad (Chart 4). It can’t do that anymore, so the government has consciously chosen to spur investment spending in order to prop up employment. Since a lot of investment spending is financed through credit, debt levels have risen. It really is just that simple.   Arthur: First, neither the stock nor the flow of credit and deposits has any relevance to (1) the economic term “savings;” (2) a country’s capital stock; or (3) national wealth, contrary to what Peter claims. China’s broad money supply (M2) now stands at 190 trillion yuan, equivalent to US$28 trillion (Chart 5, top panel). It is equal to the size of broad money supply in the U.S. and the euro area combined (US$14 trillion each). Yet, China’s nominal GDP is only two-thirds the size of the U.S. Does the level of China’s wealth and capital stock justify it having broad money supply (US$28 trillion) equivalent to the U.S. and the euro area combined? Chart 5 (Arthur)“Helicopter” Money In China Second, are Chinese households and companies willing to hold all RMB deposits that banks have created “out of thin air”? The answer: not really. Without capital controls, a notable portion of these deposits would have rushed into the foreign exchange markets and caused currency depreciation. Another sign of growing reluctance to hold the yuan is that households have been swapping their RMB deposits for real assets (property) at astronomical valuations. There is a bubble in China but people are looking for reasons to justify why it is different this time. Caroline: OK, let’s get away from the term “savings,” and agree that China continues to generate a chronic surplus of production of goods and services relative to consumption, and that how China chooses to intermediate that surplus is the most market-relevant issue. Arthur, you have used the terms “money bubble” and “helicopter money” in relation to China. This implies that banks are unconstrained in their ability to make loans. Just because savings don’t equal deposits, and banks can create deposits when they make loans doesn’t mean there is no relationship between the flow of credit and the stock of deposits. Arthur: Money supply and deposits expand only when banks originate a loan or buy an asset from a non-bank. In short, both credit and money/deposits are created by commercial banks “out of thin air.” This is true for any country.2 Consider a loan transaction by a German commercial bank. When it grants a €100 loan to a borrower, two accounting entries occur on its balance sheet. On the assets side, the amount of loans, and therefore total assets, increases by €100. Simultaneously, on the liabilities side, this accounting entry creates €100 of new deposits “out of thin air” (Figure 1). Hence, new purchasing power of €100 has been created via a simple accounting entry, which otherwise would not exist. Critically, no one needed to save for this loan and money to be originated. The bank does not transfer someone else’s deposits to the borrower; it creates a new deposit when it lends. Banks also create deposits/money “out of thin air” when they buy securities from non-banks. In China, fiscal stimulus is largely financed by commercial banks – banks purchase more than 80% of government-issued bonds. This also leads to money creation. In short, when banks originate too much credit – as they have in China – they generate a money bubble. The money bubble is the mirror image of a credit bubble. Chinese banks have created 141 trillion yuan (US$21 trillion) of new money since 2009, compared with $8.25 trillion created in the U.S., euro area, and Japan combined over the same period (Chart 5, bottom panel). This is why I refer to it as “helicopter” money. Caroline: If banks need capital and liquidity to make loans, and deposits are one potential source of funds, don't these capital and liquidity constraints drive banks’ willingness and ability to lend, creating a link between the two variables? Arthur: Let me explain how mainland banks were able to circumvent those regulatory lending constraints. In 2009, they expanded their credit assets by about 30%. Even though a non-trivial portion of those loans were not paid back, banks did not recognize NPLs and instead booked large profits. By retaining a portion of those earnings, they boosted their equity, say, by 20%. As a result, the next year they were able to expand their credit assets by another 20% and so on. If banks lend and do not recognize bad loans, they can increase their equity and continue lending. With respect to liquidity, deposits are not liquidity for banks; excess reserves at the central bank are true liquidity for them. The reason why banks need to attract deposits is not to appropriate the deposits themselves, but to gain access to the excess reserves that come with them. When a person shifts her deposit from Bank A to Bank B, the former transfers a similar amount of excess reserves (liquidity) to the latter. When expanding their credit assets aggressively, banks can: (1) create more loans per one unit of excess reserves/liquidity, i.e., expand the money multiplier; and (2) borrow excess reserves/liquidity from the central bank or other banks. Chinese banks have used both channels to expand their balance sheets over the past 10 years (Chart 6). Chart 6 (Arthur)Broad Money Can Expand Without Growing Banks' Excess Reserves At The Central Bank Crucially, commercial banks create deposits, but they cannot create excess reserves (liquidity).3 The latter are issued only by central banks “out of thin air.” So, neither deposits nor excess reserves have any relevance to household or national “savings.” Caroline: Peter, Arthur argues that Chinese credit policy has been unconstrained by the traditional metrics of capital adequacy that prevail in capitalist, free-market economies. In other words, there is no connection between the availability of funds to lend via deposits in the banking system, and the pace of credit creation. Rather, the central bank has controlled the terms and volume of lending via regulation and fiat reserve provisioning. You’ve argued that credit creation has served the greater good of propping up employment via investment spending. Moreover, you posit that countries with a surplus of production over consumption will invariably experience high levels of credit creation. Our colleague, Martin Barnes, has analyzed national savings rates (as a proxy for over-production) relative to debt-GDP ratios in other countries. The relationship doesn’t look that strong elsewhere (Chart 7). Please elaborate on why you see credit growth as an inevitable policy response to the dearth of aggregate demand we observe in China? Peter: I would not say that countries with a surplus of production over consumption will invariably experience high levels of credit creation. For example, if most business investment is financed through retained earnings, you can have a lot of investment with little new debt. Debt can also result from activities not directly linked to the intermediation of savings. For instance, if you take out a mortgage to buy some land, your consumption and savings need not change, even though debt will be created. I think Arthur and I agree on this point. Thus, I am not saying that debt is always and everywhere the result of savings. I am simply pushing back against Arthur’s extremist position that debt never has anything to do with savings. Caroline: So what determines the level of debt in an economy in your view, Peter? Peter: In general, debt levels will rise if there are large imbalances between income and spending within society and/or if there are significant differences in the mix of assets people wish to hold. Think about the U.S. in the pre-financial crisis period. First, there was a surge in income inequality beginning in the early 1980s. For all intents and purposes, rich households with excess savings ended up lending their surplus income to poor households struggling to pay their bills. Overall savings did not rise, but debt levels still increased. That’s one reason why Martin’s chart doesn’t show a strong correlation between the aggregate savings rate and debt-to-GDP. Sometimes you need to look beneath the aggregate numbers to see the savings intermediation taking place. Unlike in the U.S., even poor Chinese households are net savers (Chart 8). Thus, the aggregate savings rate in China is very high4 (Chart 9). Much of these savings are funnelled to finance investment in the corporate and public sectors. This fuels debt growth. Chart 9 (Peter)Chinese Households Have More Savings Than The U.S., Europe And Japan Combined The second thing that happened in the U.S. starting in 2000 was a massive housing boom. If you bought a second home with credit, you ended up with one more asset (the house) but one more liability (the mortgage). The person who sold you the home ended up losing one asset (the house) but gaining another asset (a bigger bank deposit). The net result was both higher debt and higher bank deposits. Lending to finance asset purchases has also been a big source of debt growth in China, as it was in the U.S. before the crisis. The U.S. mortgage boom ended in tears, and so the question that we should be asking is whether the Chinese debt boom will end the same way. Arthur: We agreed not to use the term “savings,” yet Peter again refers to “savings” being funnelled into credit. As I explained above, banks do not funnel “savings” (i.e., “excess production”) into credit. China, Japan, and Germany have high “savings” rates because they produce a lot of steel, chemicals, autos, and machinery that literally cannot be consumed and, thus, are recorded as “savings.” The U.S. produces too many services that are consumed/expensed and, hence, not recorded as “savings.” That is why the U.S. has a lower “savings” rate. Chart 10 (Arthur)The Myth Of Deficient Demand In China Economic textbook discussions on “savings” and “investment” are relevant for a barter economy where banks do not exist. When this framework is applied to modern economies with banks, it generates a lot of confusion.5 Caroline: OK, so Peter argues that an imbalance between spending and income CAN be a marker for high debt levels. Arthur, please explain why you see no relationship between China’s chronic shortfall in demand and authorities’ explicit decision to support growth via credit creation. Arthur: First, China does not have deficient demand – consumer spending and capital expenditures have been growing at 10% and 9.4%, respectively, in real terms annually compounded for the past 10 years (Chart 10). The mainland economy has been suffering from excess production, not a lack of demand. China has invested a lot (Chart 11) and ended up with too much capacity to produce steel, cement, chemicals and other materials as well as machinery and industrial goods. So, China has an excess production of goods relative to firms’ and households’ underlying demand. In a market economy, these producers would become non-profitable, halt their investments, and shut down some capacity. Chart 11 (Arthur)China Has Been Over-Investing On An Unprecedented Scale In China, to keep the producers of these unwanted goods operating, the government has allowed and encouraged banks to originate loans creating new purchasing power literally “out of thin air” to purchase these goods. This has created a credit/money bubble. In a socialist system, banks do not ask debtors to repay loans and government officials are heavily involved in resource and capital allocation. China’s credit system and a growing chunk of its economy have been operating like a typical socialist system. Socialism leads to lower productivity growth for well-known reasons. With labor force growth set to turn negative, productivity is going to be the sole source of China’s potential growth rate. If the nation continues expanding this money/credit bubble to prop up zombie enterprises, its potential growth rate will fall considerably. As the potential growth rate drops, recurring stimulus will create nominal but not real growth. In short, the outcome will be stagflation. Caroline: The theoretical macro frameworks that you have both outlined make for interesting thought experiments, and spirited intellectual debate. However, investors are most concerned about the sustainability of China’s explosive credit growth, implications for the country’s growth rate, and the return on invested capital. Arthur, given your perspective on how Chinese credit policy has been designed and implemented, please outline the contours of how and when you see the music stopping, and the debt mountain crumbling. Arthur: Not every credit bubble will burst like the U.S. one did in 2008. For example, in the case of the Japanese credit bubble, there was no acute crisis. The bubble deflated gradually for about 20 years. In the cases of the U.S. (2008), Japan (1990), the euro area (2008-2014), Spain (2008-2014) and every other credit bubble, a common adjustment was a contraction in bank loans in nominal terms (Chart 12). Chart 12 (Arthur)All Credit Booms Have Been Followed By Contracting Bank Loans (I) Chart 12 (Arthur)All Credit Booms Have Been Followed By Contracting Bank Loans (II) Why do banks stop lending? The reason is that banks’ shareholders absorb the largest losses from credit booms. Given that banks are levered at least 20-to-1 at the peak of a typical credit boom, every $1 of non-performing loans leads to a $20 drop in their equity value. Bank shareholders halt the flow of credit to protect their wealth. Chart 13 (Arthur)China: Deleveraging Has Not Yet Begun In fact, credit in China is still growing at a double-digit rate, above nominal GDP growth (Chart 13). Hence, aggregate deleveraging in China has not yet begun. If banks do not curtail credit origination, the music will not stop. However, uninterrupted credit growth happens only in a socialist system where banks subsidize the economy at the expense of their shareholders. But even then, there is no free lunch. Credit origination by banks also expands the money supply as discussed above. An expanding money bubble will heighten devaluation pressure on the yuan in the long run. The enormous amount of money supply/deposits – the money bubble – in China is like “the sword of Damocles” hanging over the nation’s currency. Chinese households and businesses are becoming reluctant to hold this ballooning supply of local currency. Continuous “helicopter” money will only increase their desire to diversify their RMB deposits into foreign currencies and assets. Yet, there is an insufficient supply of foreign currency to accommodate this conversion. The nation’s current account surplus has almost vanished while the central bank carries US$3 trillion in foreign exchange reserve representing only 11% of the yuan deposits and cash in circulation (Chart 14). It is inconceivable that China can open its capital account in the foreseeable future. “Helicopter” money also discourages innovation and breeds capital misallocation, which reduces productivity growth. A combination of slowing productivity growth, and thus potential GDP, and strong money growth ultimately lead to stagflation – the dynamics endemic to socialist systems. Peter: Arthur’s answer implicitly assumes that private investment would increase if the government removed credit/fiscal stimulus. But where is the evidence for that? We had just established that the Chinese economy suffers from a lack of aggregate demand. Public-sector spending, to the extent that it increases employment and incomes, crowds in private-sector investment rather than crowding it out. Ask yourself what would have happened if China didn’t build that “bridge to nowhere.” Would those displaced construction workers have found more productive work elsewhere or would they have remained idle? The answer is almost certainly the latter. After all, the reason the Chinese government built the bridge in the first place was to increase employment in an economy that habitually struggles to consume enough of what is produces. Arthur talks about the “misallocation” of resources. But doesn’t an unemployed worker also represent a misallocation of resources? In my view, it certainly does – and one that is much more threatening to social stability than an underutilized bridge or road. If you understand the point above, you will also understand why Arthur’s comparison between Chinese banks and say, U.S. banks is misplaced. The Chinese government is the main shareholder in Chinese banks. The government cares more about social stability than anything else. There is no way it would let credit growth plunge. Moreover, as the main shareholder, the government has a strong incentive to raise the share price of Chinese banks. After all, it is difficult to have a reserve currency that rivals the U.S. dollar, as China aspires to have, if your largest banks trade like penny stocks. My guess is that the Chinese government will shut down a few small banks to “show” that it is concerned about moral hazard, but then turn around and allow the larger banks to sell their troubled loans to state-owned asset management companies on very favourable terms (similar to what happened in the early 2000s). Once investors get wind that this is about to happen, Chinese bank shares will rally like crazy. Caroline: Isn’t shuffling debt from one sector of the economy to another akin to a shell game? Wouldn’t rampant debt growth eventually cause investors to lose confidence in the currency? Peter: China has a problem with the composition of its debt, not with its total value. Debt is a problem when the borrower can’t or won’t repay the loan. Chart 15 (Peter)China Is On Course To Lose More Than 400 Million Workers I completely agree that there is too much shadow bank lending in China. There is also too much borrowing by state-owned enterprises. Ideally, the Chinese government would move all this quasi-public spending onto its own balance sheet. It would significantly raise social spending to discourage precautionary household savings. It would also adopt generous pro-natal policies — free childcare, education, government paid parental leave, and the like. The fact that the Chinese working-age population is set to shrink by 400 million by the end of the century is a huge problem (Chart 15). If the central government borrowed and spent more, state-owned companies and local governments would not have to borrow or spend as much. Banks could then increase their holding of high-quality central government bonds. Debt sustainability is only a problem if the interest rate the government faces exceeds the growth rate of the economy.6 That is manifestly not the case in China (Chart 16). And why are interest rates so low in relation to growth? Because Chinese households save so much! We simply can’t ignore the role of savings in the discussion. Chart 16 (Peter)China: High Levels Of Household Savings Have Kept Interest Rates Below The Growth Rate Of The Economy As far as the currency is concerned, if debt growth rose so much that the economy overheated and inflation soared, then yes, the yuan would plunge. But that’s not what we are talking about here. We are talking about bringing debt growth to a level that generates just enough demand to achieve something resembling full employment. No one is calling for raising debt growth beyond that point. Curbing debt growth in a demand-deficient economy, as Arthur seems to be recommending, would cause unemployment to rise. Investors would then bet that the Chinese government would try to boost net exports by engineering a currency devaluation. Capital outflows would intensify. Far from creating the conditions for a weaker yuan, fiscal/credit stimulus obviates the need for a currency depreciation. Caroline: Peter, even if we accept your argument that the counterfactual of curbing credit growth in a demand-deficient economy would be a more deflationary outcome than sustaining the government-sponsored credit growth engine, how is building bridges to nowhere a positive sum for investors? Even if this strategy maintains social stability in the interests of the CCP’s regime preservation, won’t investors eventually recoil at the retreat to socialism that Arthur outlines, reducing the appeal of holding the yuan, even if, as you both seem to agree, no apocalyptic debt crisis is at hand? In other words, isn’t two times nothing still nothing? Peter: First of all, many of these infrastructure projects may turn out to be quite useful down the road, pardon the pun. Per capita vehicle ownership in China is only one one-fifth of what it is in the United States, and one-fourth of what it is in Japan (Chart 17). A sparsely used expressway today may be a clogged one tomorrow. Chart 17 (Peter)The Automobile Ownership Rate Is Still Quite Low In China Would China really be better off if it had fewer infrastructure projects and more big screen TVs? An economy where people are always buying stuff they don’t need, with money they don’t have, to impress people they don’t like, is hardly a recipe for success. I am not sure what these references to socialism are supposed to accomplish. You want to see a real retreat to socialism? Try creating millions of unemployed workers with no jobs and no hope. All sorts of pundits decried Franklin Roosevelt’s New Deal as creeping socialism. The truth is that the New Deal took the wind out of the sails of the fledgling U.S. communist movement at the time. Arthur: I believe that Peter is confusing the structural and cyclical needs for stimulus. When an economy is in a recession – banks are shrinking their balance sheets and property prices are deflating – the authorities must undertake fiscal and credit stimulus. Chart 18 (Arthur)What Will Productivity Growth Look Like If Public Officials Allocate 55%-60% Of GDP? Credit and fiscal stimulus made sense in China in early 2009 when growth plunged. However, over the past 10 years, we have witnessed credit and property market booms of gigantic proportions. Does this economy warrant continuous stimulus? What will productivity growth look like if government bureaucrats continuously allocate 55-60% of GDP each year (Chart 18)? Caroline: Arthur and Peter, you can both argue with one another about the semantic economic definition of the term ‘savings’, the implications of chronic excess production (relative to consumption), and the root drivers of credit growth in China long past the expiry of every BCA client’s investment horizon. Clients benefit from understanding your distinct perspectives only to the extent that they inform your outlook for markets. Will each of you now please outline how you see high levels of credit in China’s economy impacting the following over a cyclical (6-12 month) and structural (3-5 year) horizon: Global growth Commodity prices China-geared financial assets Peter: Regardless of what one thinks about the root causes of China’s high debt levels, it seems certain to me that the Chinese are going to pick up the pace of credit/fiscal stimulus over the next six months in response to slowing growth and trade war uncertainties. If anything, the incentive to open the credit spigots this time around is greater than in the past because the Chinese government wants to have a fast-growing economy to gain leverage over trade negotiations with the U.S. Chart 19 (Peter)Stronger Chinese Credit Growth Bodes Well For Commodity Prices Chart 20 (Peter)The Dollar Is A Countercyclical Currency Stronger Chinese growth will boost growth in the rest of the world. Commodity prices will rise (Chart 19). As a counter-cyclical currency, the U.S. dollar will likely peak over the next month or so and then weaken in the back half of 2019 and into 2020 (Chart 20). The combination of stronger Chinese growth, higher commodity prices, and a weaker dollar will be manna from heaven for emerging markets. If a trade truce between China and the U.S. is reached, investors should move quickly to overweight EM equities. European stocks should also benefit. Looking further out, China’s economy will slow in absolute terms. In relative terms, however, Chinese growth will remain near the top of the global rankings. China has one of the most educated workforces in the world (Chart 21). Assuming that output-per-hour reaches South Korean levels by the middle of the century, Chinese real GDP would need to expand by about 6% per year over the next decade (Chart 22). That’s a lot of growth – growth that will eventually help China outgrow its debt burden. Chart 22 (Peter)China Has More Catching Up To Do Keep in mind that credit growth of 1% when the debt-to-GDP ratio is 300% yields 3% of GDP in credit stimulus, compared with only 1% of stimulus when the debt-to-GDP ratio is 100%. That does not mean that more debt is intrinsically a good thing, but it does mean that China will eventually be able to slow debt growth even if excess savings remains a problem. Structurally, Chinese and EM equities will likely outperform their developed market peers over a 3-to-5 year horizon. The P/E ratio for EM stocks is currently 4.7 percentage points below that of developed markets, which is below its long-term average (Chart 23). While EM EPS growth has lagged DM earnings growth over the past eight years, the long-term trend still favors EM (Chart 24). EM currencies will appreciate over this period, with the RMB leading the way. Chart 23 (Peter)EM Stocks: Valuations Are Attractive Chart 24 (Peter)Earnings Growth In EM Has Outpaced That Of DM Over The Long Haul Arthur: China is facing a historic choice between two scenarios. Medium- and long-term macro outcomes will impact markets differently in each case. Table 1 shows my cyclical and structural investment recommendations for each scenario. Table 1 (Arthur)Arthur’s Recommended Investment Strategy For China-Geared Financial Assets Allowing Markets to Play A Bigger Role = Lower credit growth (deleveraging), corporate restructuring, and weaker growth (Chart 25). This is bearish for growth and financial markets in the medium term but it will make Chinese stocks and the currency structural (long-term) buys. Credit/Money Boom Persists (Socialist Put) = Secular Stagnation, Inflation and Currency Depreciation: The structural outlook is downbeat but there are mini-cycles that investors could play (Chart 26). Cyclically, China-geared financial assets still remain at risk. However, lower asset prices and more stimulus in China could put a floor under asset prices later this year. Timing these mini-cycles is critical. A buy-and-hold strategy for Chinese assets will not be appropriate in this scenario. In short, capitalism is bad but socialism is worse. I hope China will pursue the first path. Caroline: Thank you both for clarifying your perspectives. Over a multi-year horizon, markets will render the ultimate judgement on whether China’s credit boom has represented a reckless misallocation of capital, or a rational policy response to an imbalance between domestic spending and income. In the meantime, we will monitor the complexion of Chinese stimulus and evidence of its global growth multiplier effect over the coming weeks and months. These will be the key variables to watch as we determine when and at what level to upgrade BCA’s cyclical outlook for China-geared assets. Can’t wait for that debate. Footnotes 1 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Reports, “Misconceptions About China's Credit Excesses,” dated October 26, 2016 and “The True Meaning Of China's Great 'Savings' Wall,” dated December 20, 2017. 2 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Reports, “Misconceptions About China's Credit Excesses,” dated October 26, 2016 and “The True Meaning Of China's Great 'Savings' Wall,” dated December 20, 2017. 3 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Report, “China's Money Creation Redux And The RMB,” dated November 23, 2016. 4 For a discussion on the reasons behind China’s high savings rate, please see Global Investment Strategy Weekly Report, “China’s Savings Problem,” dated January 25, 2019. 5 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Report, “Is Investment Constrained By Savings? Tales Of China And Brazil,” dated March 22, 2018. 6 For a detailed discussion of these issues, please see Global Investment Strategy Weekly Report, “Is There Really Too Much Government Debt In The World?” dated February 22, 2019 and “Chinese Debt: A Contrarian View,” dated April 19, 2019.

Special Report

Dear Client, Credit in China has expanded at an exponential pace, with the country’s debt-to-GDP ratio climbing from 143% to more than 250% over the last decade. The speed and scale of China’s debt surge dwarfs Japan and the U.S.’ respective credit binges in the 1980’s and 2000’s, each of which ultimately led to financial market meltdowns. Why should China’s experience be any different? Given that China has pursued a different economic model whereby the banking sector is largely state-sponsored and the currency is tightly managed by the central bank, the answer to this pressing question for global markets is the subject of spirited debate at BCA and within the investment community at large. Clients are already aware that my colleagues, Peter Berezin and Arthur Budaghyan, disagree on the macro and market ramifications of China’s decade-long credit boom. The aim of this report is to provide visibility on the root sources of the view divergence, not to reconcile the gaps. We hope these insights will help shape your own conviction about this important topic. Caroline Miller Global Strategy Feature   Caroline: Arthur, your cautious outlook towards emerging markets in general and China’s prospects in particular stems from your belief that China’s economy is dangerously addicted to credit as a growth driver. Please explain why you dismiss the more sanguine view that China’s elevated debt burden is a function of an equally unusually high household savings rate. Arthur: It is simple: When people use the word “savings,” they typically and intuitively refer to bank deposits or securities investments; but this is incorrect. Chart 1 (Arthur)No Empirical Evidence That Deposits = 'Savings' Money supply/deposits in the banking system have no relationship with the savings rate of a nation in general or households in particular (Chart 1). When households save, they do not change the amount of money supply and deposits. Hence, households’ decision to save neither alters liquidity in the banking system nor helps banks to originate loans. In fact, banks do not intermediate deposits into loans or savings into credit.1 The terms “savings” in economics does not denote an increase in the stock of money and deposits. The term “savings” in economics means the amount of goods and services produced but not consumed. When an economy produces a steel bar, it is registered as national “savings.” We cannot consume (say, eat or expense) a steel bar. Therefore, once a steel bar or any equipment is produced, economic statistics will count it as “savings.” Besides, the sole utilization of a steel bar is in capital goods and construction, and hence, it cannot be consumed. Once a steel bar is produced, both national savings and investment will rise. That is how the “savings” = “investment” identity is derived. Chart 2 (Arthur)Chinese Households Are More Leveraged Than U.S. Ones It would avoid confusion and help everyone if economists were to call it “excess production” not “excess savings.” Banks do not need “excess production” – i.e., national “savings” – to create loans. Critically, the enormous amount of bank deposits in China is not due to household “savings” but is originated by banks “out of thin air.” In fact, Chinese households are now more leveraged than U.S. ones (Chart 2). The surge of credit and money supply in China during the past 10 years has been due to animal spirits running wild among lenders and borrowers on the mainland, not its households’ “savings.” In short, the root of China’s credit bubble is not any different from Japan’s (in the 1980s), or the U.S.’ (in the 2000s) and so on. Peter: Yes, banks can create deposits “out of thin air,” as Arthur says. However, people must be willing to hold those deposits. The amount of deposits that households and businesses wish to hold reflects many things, including the interest rate paid on deposits and the overall wealth of the society. The interest rate is a function of savings. The more people save, the lower interest rates will be. And the lower interest rates are, the more demand for credit there will be (Chart 3). It’s like asking what determines how many apples are consumed. Is it how many apple trees farmers want to plant or how many apples people want to eat? The answer is both. Prices adjust so that supply equals demand. How about national wealth? To a large extent, wealth represents the accumulation of tangible capital – factories, plant and machinery, homes and office buildings: the sort of stuff that banks can use as collateral for lending. And what determines how much tangible capital a country possesses? The answer is past savings, of the exact sort Arthur is referring to: the excess of production over consumption. So this form of “economic” savings also plays an important indirect role in determining the level of bank deposits. Chart 4 (Peter)China: From Exporting Savings To Investing Domestically And Building Up Debt I think the main problem with Arthur’s argument is that he is observing an accounting identity, which is that total bank assets (mostly loans) must equal liabilities (mostly deposits and capital) in equilibrium, without fully appreciating the economic forces – savings being one of them – that produce this equilibrium. In any case, the whole question of whether deposits create savings or savings create deposits misses the point. China’s fundamental problem is that it does not consume enough of what it produces. In the days when China had a massive current account surplus, it could export its excess savings abroad (Chart 4). It can’t do that anymore, so the government has consciously chosen to spur investment spending in order to prop up employment. Since a lot of investment spending is financed through credit, debt levels have risen. It really is just that simple.   Arthur: First, neither the stock nor the flow of credit and deposits has any relevance to (1) the economic term “savings;” (2) a country’s capital stock; or (3) national wealth, contrary to what Peter claims. China’s broad money supply (M2) now stands at 190 trillion yuan, equivalent to US$28 trillion (Chart 5, top panel). It is equal to the size of broad money supply in the U.S. and the euro area combined (US$14 trillion each). Yet, China’s nominal GDP is only two-thirds the size of the U.S. Does the level of China’s wealth and capital stock justify it having broad money supply (US$28 trillion) equivalent to the U.S. and the euro area combined? Chart 5 (Arthur)“Helicopter” Money In China Second, are Chinese households and companies willing to hold all RMB deposits that banks have created “out of thin air”? The answer: not really. Without capital controls, a notable portion of these deposits would have rushed into the foreign exchange markets and caused currency depreciation. Another sign of growing reluctance to hold the yuan is that households have been swapping their RMB deposits for real assets (property) at astronomical valuations. There is a bubble in China but people are looking for reasons to justify why it is different this time. Caroline: OK, let’s get away from the term “savings,” and agree that China continues to generate a chronic surplus of production of goods and services relative to consumption, and that how China chooses to intermediate that surplus is the most market-relevant issue. Arthur, you have used the terms “money bubble” and “helicopter money” in relation to China. This implies that banks are unconstrained in their ability to make loans. Just because savings don’t equal deposits, and banks can create deposits when they make loans doesn’t mean there is no relationship between the flow of credit and the stock of deposits. Arthur: Money supply and deposits expand only when banks originate a loan or buy an asset from a non-bank. In short, both credit and money/deposits are created by commercial banks “out of thin air.” This is true for any country.2 Consider a loan transaction by a German commercial bank. When it grants a €100 loan to a borrower, two accounting entries occur on its balance sheet. On the assets side, the amount of loans, and therefore total assets, increases by €100. Simultaneously, on the liabilities side, this accounting entry creates €100 of new deposits “out of thin air” (Figure 1). Hence, new purchasing power of €100 has been created via a simple accounting entry, which otherwise would not exist. Critically, no one needed to save for this loan and money to be originated. The bank does not transfer someone else’s deposits to the borrower; it creates a new deposit when it lends. Banks also create deposits/money “out of thin air” when they buy securities from non-banks. In China, fiscal stimulus is largely financed by commercial banks – banks purchase more than 80% of government-issued bonds. This also leads to money creation. In short, when banks originate too much credit – as they have in China – they generate a money bubble. The money bubble is the mirror image of a credit bubble. Chinese banks have created 141 trillion yuan (US$21 trillion) of new money since 2009, compared with $8.25 trillion created in the U.S., euro area, and Japan combined over the same period (Chart 5, bottom panel). This is why I refer to it as “helicopter” money. Caroline: If banks need capital and liquidity to make loans, and deposits are one potential source of funds, don't these capital and liquidity constraints drive banks’ willingness and ability to lend, creating a link between the two variables? Arthur: Let me explain how mainland banks were able to circumvent those regulatory lending constraints. In 2009, they expanded their credit assets by about 30%. Even though a non-trivial portion of those loans were not paid back, banks did not recognize NPLs and instead booked large profits. By retaining a portion of those earnings, they boosted their equity, say, by 20%. As a result, the next year they were able to expand their credit assets by another 20% and so on. If banks lend and do not recognize bad loans, they can increase their equity and continue lending. With respect to liquidity, deposits are not liquidity for banks; excess reserves at the central bank are true liquidity for them. The reason why banks need to attract deposits is not to appropriate the deposits themselves, but to gain access to the excess reserves that come with them. When a person shifts her deposit from Bank A to Bank B, the former transfers a similar amount of excess reserves (liquidity) to the latter. When expanding their credit assets aggressively, banks can: (1) create more loans per one unit of excess reserves/liquidity, i.e., expand the money multiplier; and (2) borrow excess reserves/liquidity from the central bank or other banks. Chinese banks have used both channels to expand their balance sheets over the past 10 years (Chart 6). Chart 6 (Arthur)Broad Money Can Expand Without Growing Banks' Excess Reserves At The Central Bank Crucially, commercial banks create deposits, but they cannot create excess reserves (liquidity).3 The latter are issued only by central banks “out of thin air.” So, neither deposits nor excess reserves have any relevance to household or national “savings.” Caroline: Peter, Arthur argues that Chinese credit policy has been unconstrained by the traditional metrics of capital adequacy that prevail in capitalist, free-market economies. In other words, there is no connection between the availability of funds to lend via deposits in the banking system, and the pace of credit creation. Rather, the central bank has controlled the terms and volume of lending via regulation and fiat reserve provisioning. You’ve argued that credit creation has served the greater good of propping up employment via investment spending. Moreover, you posit that countries with a surplus of production over consumption will invariably experience high levels of credit creation. Our colleague, Martin Barnes, has analyzed national savings rates (as a proxy for over-production) relative to debt-GDP ratios in other countries. The relationship doesn’t look that strong elsewhere (Chart 7). Please elaborate on why you see credit growth as an inevitable policy response to the dearth of aggregate demand we observe in China? Peter: I would not say that countries with a surplus of production over consumption will invariably experience high levels of credit creation. For example, if most business investment is financed through retained earnings, you can have a lot of investment with little new debt. Debt can also result from activities not directly linked to the intermediation of savings. For instance, if you take out a mortgage to buy some land, your consumption and savings need not change, even though debt will be created. I think Arthur and I agree on this point. Thus, I am not saying that debt is always and everywhere the result of savings. I am simply pushing back against Arthur’s extremist position that debt never has anything to do with savings. Caroline: So what determines the level of debt in an economy in your view, Peter? Peter: In general, debt levels will rise if there are large imbalances between income and spending within society and/or if there are significant differences in the mix of assets people wish to hold. Think about the U.S. in the pre-financial crisis period. First, there was a surge in income inequality beginning in the early 1980s. For all intents and purposes, rich households with excess savings ended up lending their surplus income to poor households struggling to pay their bills. Overall savings did not rise, but debt levels still increased. That’s one reason why Martin’s chart doesn’t show a strong correlation between the aggregate savings rate and debt-to-GDP. Sometimes you need to look beneath the aggregate numbers to see the savings intermediation taking place. Unlike in the U.S., even poor Chinese households are net savers (Chart 8). Thus, the aggregate savings rate in China is very high4 (Chart 9). Much of these savings are funnelled to finance investment in the corporate and public sectors. This fuels debt growth. Chart 9 (Peter)Chinese Households Have More Savings Than The U.S., Europe And Japan Combined The second thing that happened in the U.S. starting in 2000 was a massive housing boom. If you bought a second home with credit, you ended up with one more asset (the house) but one more liability (the mortgage). The person who sold you the home ended up losing one asset (the house) but gaining another asset (a bigger bank deposit). The net result was both higher debt and higher bank deposits. Lending to finance asset purchases has also been a big source of debt growth in China, as it was in the U.S. before the crisis. The U.S. mortgage boom ended in tears, and so the question that we should be asking is whether the Chinese debt boom will end the same way. Arthur: We agreed not to use the term “savings,” yet Peter again refers to “savings” being funnelled into credit. As I explained above, banks do not funnel “savings” (i.e., “excess production”) into credit. China, Japan, and Germany have high “savings” rates because they produce a lot of steel, chemicals, autos, and machinery that literally cannot be consumed and, thus, are recorded as “savings.” The U.S. produces too many services that are consumed/expensed and, hence, not recorded as “savings.” That is why the U.S. has a lower “savings” rate. Chart 10 (Arthur)The Myth Of Deficient Demand In China Economic textbook discussions on “savings” and “investment” are relevant for a barter economy where banks do not exist. When this framework is applied to modern economies with banks, it generates a lot of confusion.5 Caroline: OK, so Peter argues that an imbalance between spending and income CAN be a marker for high debt levels. Arthur, please explain why you see no relationship between China’s chronic shortfall in demand and authorities’ explicit decision to support growth via credit creation. Arthur: First, China does not have deficient demand – consumer spending and capital expenditures have been growing at 10% and 9.4%, respectively, in real terms annually compounded for the past 10 years (Chart 10). The mainland economy has been suffering from excess production, not a lack of demand. China has invested a lot (Chart 11) and ended up with too much capacity to produce steel, cement, chemicals and other materials as well as machinery and industrial goods. So, China has an excess production of goods relative to firms’ and households’ underlying demand. In a market economy, these producers would become non-profitable, halt their investments, and shut down some capacity. Chart 11 (Arthur)China Has Been Over-Investing On An Unprecedented Scale In China, to keep the producers of these unwanted goods operating, the government has allowed and encouraged banks to originate loans creating new purchasing power literally “out of thin air” to purchase these goods. This has created a credit/money bubble. In a socialist system, banks do not ask debtors to repay loans and government officials are heavily involved in resource and capital allocation. China’s credit system and a growing chunk of its economy have been operating like a typical socialist system. Socialism leads to lower productivity growth for well-known reasons. With labor force growth set to turn negative, productivity is going to be the sole source of China’s potential growth rate. If the nation continues expanding this money/credit bubble to prop up zombie enterprises, its potential growth rate will fall considerably. As the potential growth rate drops, recurring stimulus will create nominal but not real growth. In short, the outcome will be stagflation. Caroline: The theoretical macro frameworks that you have both outlined make for interesting thought experiments, and spirited intellectual debate. However, investors are most concerned about the sustainability of China’s explosive credit growth, implications for the country’s growth rate, and the return on invested capital. Arthur, given your perspective on how Chinese credit policy has been designed and implemented, please outline the contours of how and when you see the music stopping, and the debt mountain crumbling. Arthur: Not every credit bubble will burst like the U.S. one did in 2008. For example, in the case of the Japanese credit bubble, there was no acute crisis. The bubble deflated gradually for about 20 years. In the cases of the U.S. (2008), Japan (1990), the euro area (2008-2014), Spain (2008-2014) and every other credit bubble, a common adjustment was a contraction in bank loans in nominal terms (Chart 12). Chart 12 (Arthur)All Credit Booms Have Been Followed By Contracting Bank Loans (I) Chart 12 (Arthur)All Credit Booms Have Been Followed By Contracting Bank Loans (II) Why do banks stop lending? The reason is that banks’ shareholders absorb the largest losses from credit booms. Given that banks are levered at least 20-to-1 at the peak of a typical credit boom, every $1 of non-performing loans leads to a $20 drop in their equity value. Bank shareholders halt the flow of credit to protect their wealth. Chart 13 (Arthur)China: Deleveraging Has Not Yet Begun In fact, credit in China is still growing at a double-digit rate, above nominal GDP growth (Chart 13). Hence, aggregate deleveraging in China has not yet begun. If banks do not curtail credit origination, the music will not stop. However, uninterrupted credit growth happens only in a socialist system where banks subsidize the economy at the expense of their shareholders. But even then, there is no free lunch. Credit origination by banks also expands the money supply as discussed above. An expanding money bubble will heighten devaluation pressure on the yuan in the long run. The enormous amount of money supply/deposits – the money bubble – in China is like “the sword of Damocles” hanging over the nation’s currency. Chinese households and businesses are becoming reluctant to hold this ballooning supply of local currency. Continuous “helicopter” money will only increase their desire to diversify their RMB deposits into foreign currencies and assets. Yet, there is an insufficient supply of foreign currency to accommodate this conversion. The nation’s current account surplus has almost vanished while the central bank carries US$3 trillion in foreign exchange reserve representing only 11% of the yuan deposits and cash in circulation (Chart 14). It is inconceivable that China can open its capital account in the foreseeable future. “Helicopter” money also discourages innovation and breeds capital misallocation, which reduces productivity growth. A combination of slowing productivity growth, and thus potential GDP, and strong money growth ultimately lead to stagflation – the dynamics endemic to socialist systems. Peter: Arthur’s answer implicitly assumes that private investment would increase if the government removed credit/fiscal stimulus. But where is the evidence for that? We had just established that the Chinese economy suffers from a lack of aggregate demand. Public-sector spending, to the extent that it increases employment and incomes, crowds in private-sector investment rather than crowding it out. Ask yourself what would have happened if China didn’t build that “bridge to nowhere.” Would those displaced construction workers have found more productive work elsewhere or would they have remained idle? The answer is almost certainly the latter. After all, the reason the Chinese government built the bridge in the first place was to increase employment in an economy that habitually struggles to consume enough of what is produces. Arthur talks about the “misallocation” of resources. But doesn’t an unemployed worker also represent a misallocation of resources? In my view, it certainly does – and one that is much more threatening to social stability than an underutilized bridge or road. If you understand the point above, you will also understand why Arthur’s comparison between Chinese banks and say, U.S. banks is misplaced. The Chinese government is the main shareholder in Chinese banks. The government cares more about social stability than anything else. There is no way it would let credit growth plunge. Moreover, as the main shareholder, the government has a strong incentive to raise the share price of Chinese banks. After all, it is difficult to have a reserve currency that rivals the U.S. dollar, as China aspires to have, if your largest banks trade like penny stocks. My guess is that the Chinese government will shut down a few small banks to “show” that it is concerned about moral hazard, but then turn around and allow the larger banks to sell their troubled loans to state-owned asset management companies on very favourable terms (similar to what happened in the early 2000s). Once investors get wind that this is about to happen, Chinese bank shares will rally like crazy. Caroline: Isn’t shuffling debt from one sector of the economy to another akin to a shell game? Wouldn’t rampant debt growth eventually cause investors to lose confidence in the currency? Peter: China has a problem with the composition of its debt, not with its total value. Debt is a problem when the borrower can’t or won’t repay the loan. Chart 15 (Peter)China Is On Course To Lose More Than 400 Million Workers I completely agree that there is too much shadow bank lending in China. There is also too much borrowing by state-owned enterprises. Ideally, the Chinese government would move all this quasi-public spending onto its own balance sheet. It would significantly raise social spending to discourage precautionary household savings. It would also adopt generous pro-natal policies — free childcare, education, government paid parental leave, and the like. The fact that the Chinese working-age population is set to shrink by 400 million by the end of the century is a huge problem (Chart 15). If the central government borrowed and spent more, state-owned companies and local governments would not have to borrow or spend as much. Banks could then increase their holding of high-quality central government bonds. Debt sustainability is only a problem if the interest rate the government faces exceeds the growth rate of the economy.6 That is manifestly not the case in China (Chart 16). And why are interest rates so low in relation to growth? Because Chinese households save so much! We simply can’t ignore the role of savings in the discussion. Chart 16 (Peter)China: High Levels Of Household Savings Have Kept Interest Rates Below The Growth Rate Of The Economy As far as the currency is concerned, if debt growth rose so much that the economy overheated and inflation soared, then yes, the yuan would plunge. But that’s not what we are talking about here. We are talking about bringing debt growth to a level that generates just enough demand to achieve something resembling full employment. No one is calling for raising debt growth beyond that point. Curbing debt growth in a demand-deficient economy, as Arthur seems to be recommending, would cause unemployment to rise. Investors would then bet that the Chinese government would try to boost net exports by engineering a currency devaluation. Capital outflows would intensify. Far from creating the conditions for a weaker yuan, fiscal/credit stimulus obviates the need for a currency depreciation. Caroline: Peter, even if we accept your argument that the counterfactual of curbing credit growth in a demand-deficient economy would be a more deflationary outcome than sustaining the government-sponsored credit growth engine, how is building bridges to nowhere a positive sum for investors? Even if this strategy maintains social stability in the interests of the CCP’s regime preservation, won’t investors eventually recoil at the retreat to socialism that Arthur outlines, reducing the appeal of holding the yuan, even if, as you both seem to agree, no apocalyptic debt crisis is at hand? In other words, isn’t two times nothing still nothing? Peter: First of all, many of these infrastructure projects may turn out to be quite useful down the road, pardon the pun. Per capita vehicle ownership in China is only one one-fifth of what it is in the United States, and one-fourth of what it is in Japan (Chart 17). A sparsely used expressway today may be a clogged one tomorrow. Chart 17 (Peter)The Automobile Ownership Rate Is Still Quite Low In China Would China really be better off if it had fewer infrastructure projects and more big screen TVs? An economy where people are always buying stuff they don’t need, with money they don’t have, to impress people they don’t like, is hardly a recipe for success. I am not sure what these references to socialism are supposed to accomplish. You want to see a real retreat to socialism? Try creating millions of unemployed workers with no jobs and no hope. All sorts of pundits decried Franklin Roosevelt’s New Deal as creeping socialism. The truth is that the New Deal took the wind out of the sails of the fledgling U.S. communist movement at the time. Arthur: I believe that Peter is confusing the structural and cyclical needs for stimulus. When an economy is in a recession – banks are shrinking their balance sheets and property prices are deflating – the authorities must undertake fiscal and credit stimulus. Chart 18 (Arthur)What Will Productivity Growth Look Like If Public Officials Allocate 55%-60% Of GDP? Credit and fiscal stimulus made sense in China in early 2009 when growth plunged. However, over the past 10 years, we have witnessed credit and property market booms of gigantic proportions. Does this economy warrant continuous stimulus? What will productivity growth look like if government bureaucrats continuously allocate 55-60% of GDP each year (Chart 18)? Caroline: Arthur and Peter, you can both argue with one another about the semantic economic definition of the term ‘savings’, the implications of chronic excess production (relative to consumption), and the root drivers of credit growth in China long past the expiry of every BCA client’s investment horizon. Clients benefit from understanding your distinct perspectives only to the extent that they inform your outlook for markets. Will each of you now please outline how you see high levels of credit in China’s economy impacting the following over a cyclical (6-12 month) and structural (3-5 year) horizon: Global growth Commodity prices China-geared financial assets Peter: Regardless of what one thinks about the root causes of China’s high debt levels, it seems certain to me that the Chinese are going to pick up the pace of credit/fiscal stimulus over the next six months in response to slowing growth and trade war uncertainties. If anything, the incentive to open the credit spigots this time around is greater than in the past because the Chinese government wants to have a fast-growing economy to gain leverage over trade negotiations with the U.S. Chart 19 (Peter)Stronger Chinese Credit Growth Bodes Well For Commodity Prices Chart 20 (Peter)The Dollar Is A Countercyclical Currency Stronger Chinese growth will boost growth in the rest of the world. Commodity prices will rise (Chart 19). As a counter-cyclical currency, the U.S. dollar will likely peak over the next month or so and then weaken in the back half of 2019 and into 2020 (Chart 20). The combination of stronger Chinese growth, higher commodity prices, and a weaker dollar will be manna from heaven for emerging markets. If a trade truce between China and the U.S. is reached, investors should move quickly to overweight EM equities. European stocks should also benefit. Looking further out, China’s economy will slow in absolute terms. In relative terms, however, Chinese growth will remain near the top of the global rankings. China has one of the most educated workforces in the world (Chart 21). Assuming that output-per-hour reaches South Korean levels by the middle of the century, Chinese real GDP would need to expand by about 6% per year over the next decade (Chart 22). That’s a lot of growth – growth that will eventually help China outgrow its debt burden. Chart 22 (Peter)China Has More Catching Up To Do Keep in mind that credit growth of 1% when the debt-to-GDP ratio is 300% yields 3% of GDP in credit stimulus, compared with only 1% of stimulus when the debt-to-GDP ratio is 100%. That does not mean that more debt is intrinsically a good thing, but it does mean that China will eventually be able to slow debt growth even if excess savings remains a problem. Structurally, Chinese and EM equities will likely outperform their developed market peers over a 3-to-5 year horizon. The P/E ratio for EM stocks is currently 4.7 percentage points below that of developed markets, which is below its long-term average (Chart 23). While EM EPS growth has lagged DM earnings growth over the past eight years, the long-term trend still favors EM (Chart 24). EM currencies will appreciate over this period, with the RMB leading the way. Chart 23 (Peter)EM Stocks: Valuations Are Attractive Chart 24 (Peter)Earnings Growth In EM Has Outpaced That Of DM Over The Long Haul Arthur: China is facing a historic choice between two scenarios. Medium- and long-term macro outcomes will impact markets differently in each case. Table 1 shows my cyclical and structural investment recommendations for each scenario. Table 1 (Arthur)Arthur’s Recommended Investment Strategy For China-Geared Financial Assets Allowing Markets to Play A Bigger Role = Lower credit growth (deleveraging), corporate restructuring, and weaker growth (Chart 25). This is bearish for growth and financial markets in the medium term but it will make Chinese stocks and the currency structural (long-term) buys. Credit/Money Boom Persists (Socialist Put) = Secular Stagnation, Inflation and Currency Depreciation: The structural outlook is downbeat but there are mini-cycles that investors could play (Chart 26). Cyclically, China-geared financial assets still remain at risk. However, lower asset prices and more stimulus in China could put a floor under asset prices later this year. Timing these mini-cycles is critical. A buy-and-hold strategy for Chinese assets will not be appropriate in this scenario. In short, capitalism is bad but socialism is worse. I hope China will pursue the first path. Caroline: Thank you both for clarifying your perspectives. Over a multi-year horizon, markets will render the ultimate judgement on whether China’s credit boom has represented a reckless misallocation of capital, or a rational policy response to an imbalance between domestic spending and income. In the meantime, we will monitor the complexion of Chinese stimulus and evidence of its global growth multiplier effect over the coming weeks and months. These will be the key variables to watch as we determine when and at what level to upgrade BCA’s cyclical outlook for China-geared assets. Can’t wait for that debate. Footnotes 1 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Reports, “Misconceptions About China's Credit Excesses,” dated October 26, 2016 and “The True Meaning Of China's Great 'Savings' Wall,” dated December 20, 2017. 2 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Reports, “Misconceptions About China's Credit Excesses,” dated October 26, 2016 and “The True Meaning Of China's Great 'Savings' Wall,” dated December 20, 2017. 3 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Report, “China's Money Creation Redux And The RMB,” dated November 23, 2016. 4 For a discussion on the reasons behind China’s high savings rate, please see Global Investment Strategy Weekly Report, “China’s Savings Problem,” dated January 25, 2019. 5 For a detailed discussion of these issues, please see Emerging Markets Strategy Special Report, “Is Investment Constrained By Savings? Tales Of China And Brazil,” dated March 22, 2018. 6 For a detailed discussion of these issues, please see Global Investment Strategy Weekly Report, “Is There Really Too Much Government Debt In The World?” dated February 22, 2019 and “Chinese Debt: A Contrarian View,” dated April 19, 2019.   Strategy & Market Trends MacroQuant Model And Current Subjective Scores   Tactical Trades Strategic Recommendations Closed Trades

Special Report

Highlights Corporate Debt In Theory: Conventional theory holds that high levels of corporate debt pose a risk to the economy because they make the corporate sector more vulnerable to exogenous economic shocks. Corporate Debt In Practice: The conventional theory is contradicted by empirical evidence that links rapid private debt growth to negative economic outcomes, but shows no relationship between high debt levels and slow economic growth. The empirical evidence also links measures of credit market sentiment – such as corporate bond spreads – to future economic outcomes. We present an alternative theory of the corporate credit cycle that better aligns with the observed empirical results. The Current Risk: At present, the corporate debt measures that have historically been linked to weaker economic growth paint a fairly benign picture. We see no immediate risk to the U.S. economy from elevated corporate debt. Feature In our interactions with clients we are often asked whether corporate debt poses a risk to the U.S. economy. It’s easy to see why, U.S. nonfinancial corporate debt as a percent of GDP is higher than at any time since 1936 (Chart 1). Chart 1U.S. Corporate Debt: Highest Since 1936! This Special Report investigates the issue by looking at what recent academic theory and empirical evidence have to say about the relationship between corporate debt and economic growth. We then apply that evidence to today’s corporate debt situation to assess the economy’s current level of risk. We should note that this report focuses on potential risks stemming from the amount of outstanding debt, how quickly it is growing and how it is valued in financial markets. In a follow-up report, we will consider whether the ownership structure of the corporate bond market imparts additional risks to the economy and financial system. The Risk From Corporate Debt In Theory Conventional economic theory tells us that we should be concerned about elevated private sector debt because high debt makes the economy more vulnerable in the face of future shocks. Case in point, here is how the Federal Reserve’s Financial Stability Report describes the mechanism through which private sector debt impacts the economy: Excessive borrowing by businesses and households leaves them more vulnerable to distress if their incomes decline or the assets they own fall in value. In the event of such shocks, businesses and households with high debt burdens may need to cut back spending sharply, affecting the overall level of economic activity.1 This theory raises a few issues that we will consider in the remainder of this report: The theory suggests that the absolute amount of private sector debt matters more than its rate of growth. The theory suggests that elevated debt leads to a more severe economic downturn, but doesn’t necessarily cause the downturn. In other words, high debt simply makes the economy more vulnerable to exogenous shocks. The theory suggests that household debt and corporate debt are equally important. The Empirical Record Level Versus Growth While conventional theory implies that the crucial variable to monitor is the level of private sector debt, recent empirical evidence challenges this view. For example, a 2017 Bank of England paper considered a sample of 130 recessions across 26 countries and found that the rate of private debt growth matters much more.2 Please note that in the remainder of this report we define “debt growth” as the 3-year change in the debt-to-GDP ratio. Specifically, the researchers found a statistically and economically significant link between the severity of the recession – defined as the drawdown in per capita GDP – and the 3-year change in private debt-to-GDP that immediately preceded the downturn. They found no similar relationship using the level of private debt-to-GDP. In fact, the researchers found that the level of private debt to GDP only helped explain the severity of the recession when it was interacted with the rate of private debt growth. To quote from the paper: It appears that the level of credit before a recession matters for the severity of the downturn only when it is accompanied by a credit boom. By contrast, periods of fast credit growth appear to be associated with more severe recessions whether or not the level of credit is elevated.3 These findings suggest that the conventional theory presented above – that high debt levels make the private sector more vulnerable to exogenous shocks – is not the principle mechanism at work. We need an alternative theory to explain why the rate of debt growth is the more important variable to monitor. We discuss a possible alternative theory in the section titled “Toward A Better Theory” below. But for now, let’s consider the current state of the U.S. economy in light of the Bank of England’s findings. Chart 2 shows that the level of U.S. private sector debt-to-GDP is elevated compared to history. In fact, using data beginning in 1955, it was only higher in the run-up to the 2008 financial crisis. However, the second panel of Chart 2 shows that private sector debt growth is only 2.5%, a far cry from what was seen prior to the last three recessions. Chart 2Recession Watch: Private Debt Growth And Inflation We don’t mean to imply that a recession cannot occur with low private debt growth, but the track record of post-WWII U.S. recessions shows that every single one was preceded either by elevated private debt growth – 8% or above – or high inflation. At present, the U.S. economy shows very little risk on either front. Household Debt Versus Corporate Debt So far we’ve looked at private sector debt in total, i.e. we have combined household debt and nonfinancial corporate debt. This arguably masks the true instability in the U.S. economy, which is concentrated in the corporate sector. Chart 3 shows that low overall private sector debt growth of 2.5% is split between relatively quick corporate debt growth of 4.2% and household debt that is contracting at a rate of 1.8%. If we ignore the household sector’s persistent deleveraging, we see that current corporate debt growth of 4.2% is not that far below the peaks of 6.9%, 7.9% and 8% seen prior to each of the last three recessions. Chart 3U.S. Private Debt Growth Is Driven By Corporate Sector This raises two interesting questions. First, are corporate debt and household debt equally de-stabilizing for the economy? And relatedly, when tracking the U.S. economy should we focus on overall private sector debt, or should we monitor household and corporate sector debt individually? The track record of post-WWII U.S. recessions shows that every single one was preceded either by elevated private debt growth or high inflation. On the first question, we can turn back to the Bank of England paper. That paper presented the results from several regressions where the researchers looked at household debt growth and corporate debt growth individually. The results showed that elevated household debt growth and elevated corporate debt growth were both associated with more severe recessions, and with roughly equal coefficients. In the words of the researchers: Rapid credit growth continues to be an important predictor of the severity of a recession whether we look at lending to non-financial companies or to households, suggesting that the role of lending to businesses should not be ignored. Interestingly, this result stands in contrast to some other recent empirical work. Most notably, a 2016 paper by Atif Mian, Amir Sufi and Emil Verner (MSV). That paper looked at a panel of 30 countries between 1960 and 2012 and found that while higher household debt growth is associated with lower subsequent GDP growth, no such correlation is found with corporate debt.4 MSV summarize their basic result as follows: There is a significant negative correlation between changes in private debt and future output growth. Moreover, this negative correlation is entirely driven by the growth in household debt. The magnitude of the negative correlation is large, with a one standard deviation increase in the change in household debt to GDP ratio (6.2 percentage points) associated with a 2.1 percentage point lower growth rate during the subsequent three years. The main difference between the MSV methodology and that used by the Bank of England is that the MSV paper looks at GDP growth unconditional on whether there is a recession. In contrast, the Bank of England paper looks only at recessionary periods. A look back at past U.S. recessions makes us reluctant to ignore corporate debt growth completely. Table 1 lists every post-WWII U.S. recession, showing the peak-to-trough drawdown in GDP as a measure of the recession’s severity along with prior peaks in private debt growth, household debt growth, corporate debt growth and inflation. Table 1A History Of Post-WWII U.S. Recessions Table 1 confirms what we already stated above, that every post-WWII U.S. recession has been preceded by either rapid private sector debt growth or high inflation. If we dig deeper and look at the breakdown between household debt growth and corporate debt growth we find that there have only been two recessions where peak corporate debt growth exceeded peak household debt growth. Current corporate debt growth of 4.2% is not that far below the peaks of 6.9%, 7.9% and 8% seen prior to each of the last three recessions. The first such recession occurred in 1973-75, but that recession was clearly driven by high inflation. Both household and corporate debt growth were quite low during that period. The second example is the 2001 recession. Private debt growth was elevated prior to the 2001 recession, and more heavily concentrated in the corporate sector. However, it’s important to note that the 2001 recession was also the mildest post-WWII U.S. recession. Main Takeaways We draw several conclusions from our review of the empirical research: First, we should pay attention to the rate of growth in private debt-to-GDP and downplay the level of private debt-to-GDP. The latter has very little predictive power on its own. Second, a U.S. recession is unlikely to occur in the absence of elevated private sector debt growth (above ~8%) or high inflation. At the moment, neither factor suggests that the U.S. economy is on the cusp of a downturn. Third, we should not ignore corporate debt growth. However, the MSV research suggests it might be less economically important than household debt growth. Further, the Bank of England paper shows that the severity of any future downturn is equally sensitive to both household and corporate debt, suggesting that it is reasonable to combine the two and use overall private sector debt growth as our key metric when assessing risks to the economy. Finally, the empirical research suggests that the theory of how corporate debt relates to the economy that was presented in the first section of this report is at best incomplete. That theory cannot explain why the rate of debt growth is associated with weaker economic activity, but the level of debt is not. Fortunately, some recent research proposes a few alternative theories that better align with the empirical results. These theories also suggest a few other measures of corporate credit risk that are important for investors to monitor. Looking Beyond Debt Growth So far we have focused on the difference between the level of corporate debt and the rate of corporate debt growth, but recent empirical research has also linked several other measures of ebullient credit market sentiment to future slow-downs in economic activity. Assessing Credit Market Sentiment For example, a 2016 paper by David Lopez-Salido, Jeremy Stein and Egon Zakrajsek (LSZ) shows, using U.S. data from 1929 to 2013, that “when corporate bond spreads are narrow relative to their historical norms and when the share of high-yield bond issuance in total corporate bond issuance is elevated, this forecasts a substantial slowing of growth in real GDP, business investment, and employment over the subsequent few years. Thus buoyant credit-market sentiment today is associated with a significant weakening of real economic outcomes over a medium-term horizon.”5 Before getting into the possible reason for this finding, let’s quickly look at how the U.S. economy stacks up with regard to credit market sentiment. First, the spread between Baa-rated corporate bonds and the 10-year U.S. Treasury yield – the spread measure used in the LSZ paper – is slightly above its historical average, and does not look stretched compared to history (Chart 4). Chart 4U.S. Credit Spreads Aren't Stretched Second, even a more conventional spread measure like the average option-adjusted spread from the Bloomberg Barclays Investment Grade Corporate Bond index remains fairly wide (Chart 5). Chart 5Junk Share Of New Issuance Is Falling Third, the high-yield share of new corporate bond issuance was elevated early in the recovery, especially compared to last cycle, but has declined in recent years (Chart 5, panel 2). Relatedly, the par value of outstanding junk debt as a proportion of the total par value of corporate debt has been falling since 2015 (Chart 5, bottom panel). Does Elevated Credit Market Sentiment Cause Slower Economic Growth? Of course, the empirical finding that tight credit spreads predict slower economic growth could simply reflect the fact that credit spreads respond to swings in the economic data. If our goal is to forecast economic growth, then this would suggest that we don’t need to pay much attention to credit spreads, because they are simply reflecting swings in the economy rather than causing them. However, the empirical evidence increasingly suggests that there is a causal mechanism at play. To test this, the LSZ paper employs a two-step regression procedure. In the first step, researchers model the future change in credit spreads based on the lagged level of credit spreads and the junk share of new issuance. In the second step, they use the fitted value from the first regression to predict changes in economic activity. The fact that the fitted value is significantly related to changes in economic activity implies that there is some predictable mean reversion in credit market sentiment, unrelated to economic fundamentals, that actually exerts an influence on future economic growth. LSZ suggest the following causal mechanism: Heightened levels of sentiment in credit markets today portend bad news for future economic activity. This is because mean reversion implies that when sentiment is unusually positive today, it is likely to deteriorate in the future. Moreover, a sentiment-driven widening of credit spreads amounts to a reduction in the supply of credit, especially to lower credit-quality firms. It is this reduction in credit supply that exerts a negative influence on economic activity. It follows from this analysis that if we could show that corporate bond spreads are tight relative to their “economic fair value”, then the economy would be at even greater risk from a mean reversion in credit market sentiment. While it’s difficult to identify a true “fair value” for credit spreads, Simon Gilchrist and Egon Zakrajsek (GZ) have calculated an Excess Bond Premium that measures the excess spread available in a sample of corporate bonds after removing a bottom-up estimate of expected default losses.6 Expected default losses are estimated using the Merton model and each firm’s market value of equity and face value of debt.7 Using this new measure, GZ find that “over the past four decades, the predictive power of credit spreads for economic downturns is due entirely to the Excess Bond Premium”. This stunning result is the most compelling evidence yet that swings in credit market sentiment actually cause shifts in economic activity, rather than simply reflect them. Looking at the GZ Excess Bond Premium today, we see that while it had been negative for most of the current cycle, it recently ticked above zero and has yet to recover (Chart 6). For the time being, there is no evidence of excessively optimistic credit market sentiment. Chart 6U.S. Credit Spreads Are High Relative To Fundamentals Toward A Better Theory So far we’ve seen that rapid debt growth is a better predictor of future economic weakness than high debt levels. We’ve also seen evidence that optimistic credit market sentiment (tight credit spreads, especially relative to fundamentals, and an elevated junk share of new issuance) forecasts, and likely causes, future economic weakness. Clearly, we need a better theory for why corporate debt matters for the economy than the one provided by the Federal Reserve in the first section of this report. In our view, the theory that most closely aligns with the empirical data is Nicola Gennaioli and Andrei Shleifer’s theory of Diagnostic Expectations, as detailed in their 2018 book A Crisis Of Beliefs.8 In the book, the author’s demonstrate how investors systematically overreact to new economic information. A tendency that makes forecast errors highly predictable. For example, Chart 7 shows that forecasts for what the Baa/Treasury spread will be in one year’s time are tightly linked with today’s actual spread. This means that investors inevitably expect too much future spread widening when spreads are high, and too much future tightening when spreads are low. Chart 7Forecast Errors Are Predictable Gennaioli and Shleifer integrate this systematic behavioral bias into a model that, from our perspective, better aligns with the empirical data on the relationship between corporate debt and the real economy. According to Gennaioli and Shleifer: Good economic news […] makes right-tail outcomes representative. This leads investors to both overestimate average future conditions and to neglect the unrepresentative downside risk, causing overexpansion of both leverage and real investment. When good news stops coming, investors revise their expectations down, even without adverse shocks. These revisions cause credit spreads to revert, the lenders to perform poorly, and economic and financial conditions to deteriorate, leading to deleveraging and cuts in real investment. A severe crisis occurs if arriving news is sufficiently bad as to render left-tail outcomes representative and hence overstated. This theory would seem to explain all of the key empirical findings. Investors form their expectations based on an overreaction to recent news. During an economic recovery this causes credit spreads to tighten and debt to grow rapidly. Eventually, investors realize that expectations have become unrealistically optimistic, credit spreads mean-revert and debt growth plunges. Crucially, in this model a severe economic shock is not required for credit spreads to mean-revert, only a lack of further good news to confirm investor over-optimism. Based on this theory, if we are concerned about the impact of corporate debt on the real economy we should predominantly track measures of credit market sentiment and the rate of debt growth. The theory helps reveal why the level of corporate debt has little informational value. Concluding Thoughts Conventional theory tells us that high corporate debt levels could pose a risk to the economy because they make the corporate sector more vulnerable in the face of exogenous economic shocks. However, empirical evidence suggests that this theory is of little practical value. A better theory is one where investors and corporate managers overreact to positive economic news, leading to overvaluation in credit markets and rapid debt growth. Then, when sentiment is revealed to be overly optimistic, it leads to a mean-reversion in credit spreads and a tightening of credit supply that actually causes a period of weaker economic growth. Investors inevitably expect too much future spread widening when spreads are high, and too much future tightening when spreads are low. It follows from this theory that if we are concerned about the impact of corporate debt on the real economy we should predominantly track debt growth and measures of credit market sentiment such as credit spreads and the junk share of new issuance. The U.S. economy currently looks quite stable by these measures. Overall private sector debt growth is only 2.5%. Historically, it has been above 8% prior to recessions that weren’t caused by high inflation. The GZ Excess Bond Premium also shows that credit market sentiment is not currently stretched relative to fundamentals. Ryan Swift, U.S. Bond Strategist rswift@bcaresearch.com   Footnotes 1      https://www.federalreserve.gov/publications/files/financial-stability-report-201811.pdf 2      https://www.bankofengland.co.uk/working-paper/2017/down-in-the-slumps-t… 3      Please note that the Bank of England paper uses the term “credit” in place of “debt”. In this report we use both terms interchangeably. 4      https://chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=1050&co… 5      https://www.nber.org/papers/w21879 6      https://www.federalreserve.gov/econresdata/notes/feds-notes/2016/recession-risk-and-the-excess-bond-premium-20160408.html 7      Merton, Robert C., “On The Pricing Of Corporate Debt: The Risk Structure of Interest Rates”, The Journal of Finance, Vol. 29, No. 2, May 1974. 8      Nicola Gennaioli and Andrei Shleifer, A Crisis Of Beliefs: Investor Psychology And Financial Fragility, Princeton University Press, 2018.

Highlights Chinese credit origination surpassed expectations in March. Credit growth is now clearly trending higher, and the latest data suggest that economic activity is rebounding. This bodes well for global growth. The conventional wisdom is that China’s releveraging efforts represent “short-term gain for long-term pain.” We disagree. For the most part, Chinese releveraging is inevitable, desirable, and sustainable. Credit growth is inevitable because rising debt is necessary for transforming the country’s copious savings into fixed-asset investment. It is desirable for ensuring that GDP growth stays close to trend. It is broadly sustainable because the interest rate at which the government and much of the private sector are able to borrow is well below the economy’s growth rate. In fact, under a plausible set of assumptions, faster credit growth in China could lead to a lower debt-to-GDP ratio. Stronger global growth later this year should weaken the U.S. dollar. We are closing our long DXY trade for a carry-adjusted gain of 16.4% and exiting our long USD/CNY trade for a loss of 3.1%. We are also taking profits on our short AUD/CAD, short EUR/CAD, and short EUR/RUB trades of 1.6%, 3.9%, and 8.6%, respectively, and initiating two new currency trades: short USD/RUB and long EUR/JPY. The combination of a weaker dollar and faster Chinese growth should benefit EM and European stocks. Gold hit our limit buy order of $1275/ounce and we are now long the yellow metal. Feature A Blockbuster Month For Chinese Credit Growth After turning cautious for about six months, we moved back to being bullish on global equities in late December. We also sold our put on the EEM ETF on January 3rd for a gain of 104% in anticipation of a wave of Chinese credit stimulus. Credit growth blew past expectations in January, but surprised on the downside in February. This made the March release particularly important. In the end, the March data did not disappoint those who were hoping for a solid reading. New CNY loans rose by RMB 1690 billion, above Bloomberg consensus estimates of RMB 1250 billion. Our adjusted aggregate financing measure, which excludes a number of items such as equity financing but includes local government bond issuance, rose by 12.3% year-over-year, up from 11.6% in February (Chart 1). China’s credit impulse leads the import component of its manufacturing PMI (Chart 2). The credit impulse bottomed in November 2018, which should feed into higher imports over the coming months. This week’s release of better-than-expected data on industrial production, retail sales, and housing activity all suggest that the rebound in Chinese growth is already afoot. Chart 1Chinese Credit Growth Is Rebounding... Chart 2...Which Should Bode Well For Global Exports To China   Short-Term Gain For Long-Term Pain? At times like these, the bears are always ready with their standby argument: Sure, China may be stimulating, but all that credit growth will just make the debt bubble even bigger. Once the bubble bursts, there will be hell to pay. Long-term investors should steer clear of any growth-sensitive assets. It is a seductive argument. But it is wrong. Chinese releveraging is: 1) inevitable; 2) desirable; and 3) sustainable. The fundamental macroeconomic problem that China faces is that it consumes too little of what it produces. 1. Chinese Debt Growth Is Inevitable The fundamental macroeconomic problem that China faces is that it consumes too little of what it produces. The result is a national savings rate of 45%, by far the highest of any major economy (Chart 3). Chart 3China Still Saving A Lot Chart 4From Exporting Savings To Investing Domestically And Building Up Debt   There was a time when China was able to export a large part of its excess production. Its current account surplus reached nearly 10% of GDP in 2007. As its economy has grown in relation to the rest of the world, running massive trade surpluses has become more difficult. This is especially true today, when the country is being targeted by the Trump administration and much of the international community for alleged unfair trade practices. As China’s ability to churn out large current account surpluses declined, the government moved to Plan B: propping up growth by recycling the country’s copious savings into fixed-asset investment. This process saw households park their savings in banks and other financial institutions which, in turn, lent the money out to companies and local governments in order to finance various investment projects. Not surprisingly, debt levels exploded higher (Chart 4). As China’s population ages and more workers leave the labor force, savings will decline. However, this is likely to be a slow process. In the meantime, further debt growth is inevitable. 2. Chinese Debt Growth Is Desirable In an ideal world, Chinese households would consume more of their incomes, leaving only enough savings to finance high-quality private and public investment projects. That is not the world we are living in. In a far-from-ideal world, we need to think about second-best solutions. Yes, a sizable share of Chinese investment spending goes towards projects of dubious value. Yet, the same could have been said about Japan’s fabled “bridges to nowhere.” One may regard the construction of a seldom-used bridge as a misallocation of capital. But what is the counterfactual? If the bridge had not been built, would the workers have found productive work? If not, then there also would have been a misallocation of capital – human capital – which is arguably a much more serious problem. In any case, keep in mind that the rate of return on private investment depends on the state of the economy. If an economy is suffering from chronic lack of demand, only the most worthwhile projects will be undertaken. As the economic outlook improves, the set of viable projects will expand. It is only when all excess private-sector savings have been depleted, and interest rates are rising, that public spending starts to crowd out private investment. 3. Chinese Debt Growth Is Sustainable Even if one accepts the proposition that China needs continued debt growth to maintain full employment, is it still possible that all this additional debt will push the economy into a full-blown debt crisis? Most self-professed “serious-minded” observers would say yes. But then again, many of these same observers were predicting that Japan was heading for a debt crisis when government debt reached 100% of GDP in the late 1990s. Today, Japan’s government debt-to-GDP ratio stands at about 240% of GDP, and yet interest rates remain at rock-bottom levels. China will avoid a debt crisis for the same reason Japan has been able to avoid one. Much of China’s debt stock is composed of state-owned enterprise, local government, and other forms of quasi-public sector debt. Credit policy in China is often indistinguishable from fiscal policy. Given the abundant supply of savings in the economy, most of this debt can be internally financed at fairly low interest rates. The standard equation for government debt dynamics says that the change in the debt-to-GDP ratio, D/Y, can be expressed as:1 G - T is the primary budget deficit, r  is the borrowing rate, and g is the growth rate of the economy (it is irrelevant whether r and g are defined in nominal or real terms, as long as they are both expressed the same way). China will avoid a debt crisis for the same reason Japan has been able to avoid one. The Chinese 10-year government bond yield is currently four percentage points below projected GDP growth over the next decade, which is one of the biggest gaps among the major economies (Chart 5). Arithmetically, this means that China can have as large a primary fiscal deficit as it wants. As long as r remains below g, the debt-to-GDP ratio will converge to a stable level. Chart 6 shows this point analytically. In fact, it is possible that a permanently larger budget deficit could lead to a decline in the equilibrium debt-to-GDP ratio. How could that be? The answer is revealed by the equation above. If the debt-to-GDP ratio is fairly high to begin with and an increase in the primary budget deficit leads to higher inflation (and hence, lower real rates and/or faster nominal GDP growth), this could more than fully counteract the increase in the deficit. Chart 7Stronger Growth Coincided With Accelerating Inflation And Lower Real Rates This is not just a theoretical curiosity. Historically, Chinese inflation has risen while real rates have fallen whenever GDP growth has accelerated (Chart 7). Given China’s high debt levels, even a modest amount of additional inflation could put significant downward pressure on the debt-to-GDP ratio.2  Of course, all this is predicated on the assumption that faster credit growth will not cause interest rates to rise above the growth rate of the economy. For the portion of China’s debt stock that is either directly or indirectly backstopped by the central government, this seems like a safe assumption. After all, if credit/fiscal stimulus is simply being undertaken in response to inadequate demand, there is no need for policymakers to hike rates. Things get trickier when we look at private debt. In the past, the government has encouraged state-owned banks to roll over souring loans for fear that a wave of defaults would undermine the economy and endanger social stability. More recently, however, policymakers have been backing away from this strategy due to the well-founded view that it encourages moral hazard. Faster growth in China in the second half of this year will lift Chinese imports. This will be welcome news for the rest of the world. We expect the authorities to continue taking steps to instill market discipline by allowing failing firms to, well, fail. Realistically, however, the transition to a full market-based economy will take quite a bit of time. In the interim, the government will keep cutting taxes and increasing on-budget spending in order to ensure that any decline in employment among failing firms is offset by employment growth elsewhere. In such an environment, neither a debt crisis nor a deep economic slowdown appear likely. Investment Conclusions Faster growth in China in the second half of this year will lift Chinese imports. This will be welcome news for the rest of the world. Chart 9Germany Welcomes The Upturn In Chinese Credit Growth While the U.S. will benefit from a revival in Chinese growth, Europe will gain even more (Chart 8). Germany, in particular, should see a pronounced acceleration in growth. China’s credit impulse leads Chinese automobile spending which, in turn, reliably leads euro area automobile exports, as well as overall exports (Chart 9). The recent rebound in the expectations component of the German ZEW index, as well as in the manufacturing output component of the April flash PMI, suggests that green shoots are starting to sprout (Chart 10). Italy should also benefit from the steep drop in bond yields since last October (Chart 11). Italian industrial production strongly surprised to the upside in February, suggesting that the euro area’s third biggest economy may have finally turned the corner. Chart 10Tentative Green Shoots Out Of Germany Chart 11Italy: The Drop In Bond Yields Should Boost The EconomyThe ECB will not hike rates this year even if growth shifts into higher gear, but the market will probably price in a bit more monetary tightening in 2020 and 2021. This should help lift the euro. We recommend that investors position themselves for this by going long EUR/JPY. Relatedly, we are closing our short EUR/CAD trade for a gain of 3.9%.   The U.S. dollar tends to be a countercyclical currency, meaning that it moves in the opposite direction of the global business cycle (Chart 12). This countercyclicality stems from the fact that the U.S. is more geared towards services than manufacturing compared with most other economies (Chart 13). As such, when global growth accelerates, capital tends to flow from the U.S. to the rest of the world, translating into more demand for foreign currency and less demand for dollars. Chart 12The Dollar Is A Countercyclical Currency Chart 13The U.S. Is A Low-Beta Play On Global Growth A “patient” Fed and the prospect of stronger global growth in the second half of this year are bearish for the dollar. As such, we are closing our long DXY trade for a carry-adjusted gain of 16.4% and exiting our long USD/CNY trade for a loss of 3.1%. We are also closing our short AUD/CAD trade for a gain of 1.6%. Faster Chinese growth will boost metal prices, which is bullish for the Aussie dollar. Lastly, we are switching our short EUR/RUB trade (which is currently up 8.6%) into a short USD/RUB trade. A weaker greenback and stronger global growth will be manna from heaven for international stocks, especially when priced in U.S. dollars. Investors should prepare to move European and EM equities to overweight within a global equity portfolio during the coming weeks. A “patient” Fed and the prospect of stronger global growth in the second half of this year are bearish for the dollar. We are less keen on upgrading Japanese equities. While Japanese exporters will benefit from stronger Chinese growth, the domestic economy will be weighed down by the upcoming hike in the sales tax, which is slated to take place in October. Moreover, the yen is likely to experience headwinds as global bond yields rise in relation to JGB yields. Investors contemplating buying Japanese stocks should hedge any currency risk. Finally, the price of gold fell to $1275/ounce earlier this week, triggering our buy order. With the Fed on pause, the U.S. economy starting to overheat, and the dollar likely to trend lower, bullion could shine over the coming months.   Peter Berezin, Chief Global Strategist Global Investment Strategy peterb@bcaresearch.com Footnotes 1      Please see Global Investment Strategy Weekly Report, “Is There Really Too Much Government Debt In The World?” dated February 22, 2019, for a fuller discussion of this debt sustainability equation. Strategy & Market Trends MacroQuant Model And Current Subjective Scores Tactical Trades Strategic Recommendations Closed Trades

Feature For a decade, mainstream economics has prescribed remedies for sluggish growth in the euro area on the basis of three articles of blind faith. First, that the ailment arises from structural impediments to growth; second, that in response to an ailing economy, ultra-loose monetary policy is always and everywhere effective; and third, that ‘Keynesian’ government stimuluses are at best a necessary evil and at worst a recipe for disaster. As a result, European policymakers have expended much time and energy attempting structural reforms, experimenting with ultra-loose monetary policy, while shirking government borrowing and spending. But have policymakers misdiagnosed the ailment? Chart of the WeekItaly’s Private Sector Is Paying Back Debt Why The Focus On Public Deficits And Debt Might Be Misplaced We frown upon government deficits. They are associated with crowding out and misallocation of resources. But when the private sector is running a financial surplus, the exact opposite is true. Government borrowing and spending causes no crowding out because the government is simply utilising the private sector’s surplus savings and debt repayments. And importantly, this deficit spending prevents a deflationary shrinkage of the broad money supply. Most people are aware of the size of government deficits. Few people are aware of the size of private sector surpluses; and the leakage from the national income stream that they create. By not making this connection, people might believe that government deficits are profligate. But if the private sector as a whole has a financial surplus, it makes sense for the government to borrow to support economic growth. In a similar vein, an economy’s debt sustainability depends on its total indebtedness, not on its public indebtedness or its private indebtedness in isolation. Debt becomes unsustainable when the marginal extra euro of debt results in misallocation of resources and mal-investment. At this point, the extra debt adds nothing to growth or, worse, it subtracts from growth. This is also the point at which lenders tend to be unwilling to provide the marginal loan. Therefore, debt reaches its sustainable limit when the economy has exhausted all productive uses for it. Deficit spending can prevent a deflationary shrinkage of the broad money supply. It does not matter whether these productive uses are funded with private debt or with public debt. For example, successful economies require investment in high-quality healthcare and education. Some economies fund this with private debt, while others fund it with public debt. This means that if productive private indebtedness is low, there is more scope for productive public indebtedness. Many people believe that Italy has one of the world’s most indebted economies. But this belief is wrong. Although Italy’s public indebtedness is high, Italy’s private indebtedness is one of the lowest in the world, making Italy’s total indebtedness less than that of France and the U.K., and broadly equal to that of the U.S. (Chart I-2-I-5). Crucially, Italy’s extremely low private indebtedness means that it could afford relatively high public indebtedness before reaching the limit of debt sustainability. Chart I-2Italy: Total Debt = 250% Of GDP Chart I-3France: Total Debt = 315% Of GDP Chart I-4U.K.: Total Debt = 280% Of GDP Chart I-5U.S: Total Debt = 250% Of GDP   Italy And Japan: Compare And Contrast In a normal world, the task of ensuring that private sector savings are borrowed and spent falls on the banks, which take in the savings and debt repayments and lend them out to others in the private sector who can make the best use of the funds. But if a dysfunctional banking system fails this task, the savings generated by the private sector will find no borrowers. The unrecycled funds become a leakage from the national income stream generating a persistent deflationary headwind for the economy. Welcome to Italy! Since 2008, the stock of loans to Italian households and firms has been stagnant while in real terms it has fallen (Chart of the Week). The upshot is that the real money supply has shrunk despite low private sector indebtedness, low interest rates and massive injections of ECB liquidity into the banking system. Japan’s public sector levering has been counterbalancing its private sector de-levering. After the 2008 global financial crisis Italian banks’ balance sheets were left unrepaired and undercapitalized. For an individual bank whose solvency is impaired, the right thing to do is shrink its loan book relative to its equity capital. But when the entire banking system is doing this simultaneously, the economy falls into a massive fallacy of composition: what is right for an individual bank becomes very deflationary when all banks are doing it together. Under these circumstances, an agent outside the fallacy of composition – namely, the government – must counter this deflationary headwind by borrowing and spending the un-recycled private sector savings. Welcome to Japan! The Japanese government has been doing precisely this for the past 25 years. Many people fret about the Japanese government’s persistent deficits and its ballooning public debt. What these people do not realise is that these persistent deficits are simply counterbalancing private sector de-levering. Hence, Japan’s all-important total (public plus private) indebtedness as a share of GDP has not been rising (Chart I-6). In Italy, the banking system has been dysfunctional for over a decade, preventing the private sector from borrowing (Chart I-7). Under these circumstances, the Italian government could borrow the private sector’s excess savings and debt repayments and put them to highly productive use, just like in Japan. Chart I-6Japan’s Persistent Deficits Have Been Counterbalancing Private Sector De-levering Chart I-7The Italian Banking System Has Been Dysfunctional Japan and Italy have quite similar demographics, but there is also a big difference. Despite the Japanese government’s persistent deficit and ballooning debt, the 10-year Japanese government bond seems not the slightest bit concerned and is yielding zero. Whereas in Italy, where the government finances are close to structural balance, the merest hint of a Keynesian stimulus sent the 10-year BTP yield rocketing towards 4 percent. Why? The answer is that Italy does not have its own central bank. The Japanese government bond yield is a direct function of the BoJ’s expected monetary policy. But the Italian BTP yield has two components: the ECB’s expected monetary policy plus a risk-premium for currency redenomination in the event that Italy left the euro. Italy’s problem is that even if modest deficit spending was the right policy, it would take time to prove. Meanwhile, bond vigilantes shoot first and ask questions later. The euro debt crisis was essentially a fear of currency redenomination which resulted from bond vigilantes running amok. When bond markets refuse to lend to sovereigns at a rational interest rate, maturing debt has to be refinanced at a penalising interest rate, causing an undeserved deterioration in the government’s finances. Thereby, the fear of redenomination could become a self-fulfilling prophecy. In Italy, the banking system has been dysfunctional for over a decade. The bottom line is that every economy has its own ‘tipping-point’ interest rate, at which its debt financing can flip from stability to instability. But we believe this interest rate is low everywhere. Modern Monetary Theory Simplified Modern Monetary Theory (MMT) is a hot topic of the moment. Our view is that its breakthrough is to establish the ‘appropriate’ public sector deficits in the context of private sector surpluses, and it simplifies to this question: In highly indebted economies, what is the interest rate needed to keep total (public plus private) indebtedness as a share of GDP stable, and prevent a deflationary shrinkage of the broad money supply? The answer differs slightly from economy to economy because private sector indebtedness is modestly rising in some places, stable in a few, while declining in others (Chart I-8).  But crucially, at a global level, total indebtedness is stabilising with the global bond yield within a historically depressed sideways channel (Chart I-9). Chart I-8Private Sector Indebtedness Is Not Rising As A Whole Chart I-9The Global Long Bond Yield Has Been In A Sideways Channel Admittedly, the global bond yield is now at the bottom of this channel. This means that from a tactical perspective, we can expect 10-year yields to go up about 50 bps before hitting the top of the channel. However, from a structural perspective, the interest rate needed to stabilise total indebtedness as a share of GDP now appears to be extremely low. And this means that structurally low bond yields are here to stay. Finally, I am excited to report that two of the main commentators on MMT – Richard Koo and Stephanie Kelton – are keynote speakers at our annual conference on September 26-27 in New York City. Suffice to say it will be an event not to be missed! Fractal Trading System* There are no new trades this week, leaving five open positions. 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-10 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. *  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 Dhaval Joshi, Chief European Investment Strategist dhaval@bcaresearch.com Fractal Trading System Recommendations Asset Allocation Equity Regional and Country Allocation Equity Sector Allocation Bond and Interest Rate Allocation Currency and Other Allocation Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields   Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields   Interest Rate Chart II-5Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch - Interest Rate Expectations Chart II-7Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch - Interest Rate Expectations  

Special Report

Feature This week, instead of our regular Weekly Report, we will answer clients’ most frequently asked questions (FAQs) from our recent marketing trip to the old continent. Table 1 lists these questions and below we will attempt to weave a cohesive piece and answer all of these interesting questions. Clients inquiring about “how is everyone else positioned” or the related “what is the general investor sentiment like” is by far the most FAQ we always get from the road and we purposefully omit it from Table 1. Table 1Most FAQs From The Road During our last three developed markets (DM) trips, while we cannot comment on the positioning question, with regard to general investor sentiment, Australia and New Zealand are off the charts bullish. On the opposite end of the spectrum, Europe is extremely bearish, especially continental Europe. The U.S. is somewhere in the middle. Chart 1Fed’s Pivot On Display With that out of the way, the recent broadening out of the U.S. yield curve inversion to the 10/fed funds rate took center stage in our client interactions, especially the implications of the inversion for sector positioning and the duration of the business cycle. To set the record straight, a yield curve inversion does not forecast recession. Instead, it explicitly signals that the market expects the Fed’s next move to be an interest rate cut (top panel, Chart 1). In that context, the yield curve has never had a false-positive reading. Even in May 1998, it accurately forecast that the Fed would decrease the fed funds rate as it actually did in the fallout of the LTCM meltdown later that year (bottom panel, Chart 1). As equity investors, what consumes us is the SPX’s performance following the yield curve inversion. On that front, mid-December last year we showed the results of our research and made a simple observation that the yield curve inversion almost always takes place prior to the S&P peak (Table 2, Charts 2 & 3). Table 2Yield Curve Inversions And S&P 500 Peaks Chart 3…And Then The SPX Peaks In addition, today we show the S&P 500’s return and the sector returns from the time the 10/2 yield curve slope inverts until the S&P peaks, and we summarize the results in Table 3. Table 3Sector Returns From Y/C Inversion To SPX Peak While every cycle is different, clearly it pays to have energy exposure more often than not. In contrast, high-yielding defensive sectors like utilities and telecom services fare poorly in these late-cycle iterations. Meanwhile, Table 4 highlights sector performance from the SPX peak until the U.S. recession hits. We first showed these results on May 22, 2018, and we are on track to publish a Special Report on May 5 on how to position portfolios at the onset of a Fed easing cycle, so stay tuned. Table 4Defensive Stocks Beat Late Investors remain infatuated with the recession signal that the yield curve inversion emits. Moreover, recent news of an onslaught of Unicorn IPOs that would bring stock supply to the equity market, near the $100bn mark on an annualized basis according to some estimates, have also brought forward recession fears, as smart money is cashing in on their investments. Chart 4 shows that $100bn per annum in IPOs has coincided with the SPX peak in the previous two cycles. Our long-held view remains that either a mega M&A deal in the tech or biotech space or Uber’s IPO at a stratospheric valuation could serve as the anecdote that confirms the current cycle’s peak. On the yield curve front specifically, the top panel of Chart 5 shows that the most important yield curve, the 10/2, has not yet inverted. Moreover, the 30/10 and the 30/5 slopes are steepening. True, we are late cycle, but we need all the slopes to invert to get a confirmation that the recession is a foregone conclusion. Chart 4Mind The Excess Supply Chart 510/2 Y/C Has Yet To Invert The Fed’s tightening cycle has not only inverted most parts of the yield curve starting early last December, but has inflicted some damage on profit margins. Following up from our recent profit margin work highlighting nil corporate pricing power at a time when wage costs are perking up, BCA’s Monetary Indicator signals more SPX margin pain in the coming months (Chart 6). In fact, sell-side estimates call for another three consecutive quarters of a year-over-year contraction in profit margins. Chart 6Margin Trouble In more detail, the earnings deceleration that commenced in Q4 2018 and is gaining steam is disconcerting. As a reminder, Q4 included the lower corporate tax rate and the Q/Q deceleration is not solely due to the tech sector profit warnings. Eight out of the 11 GICS1 sectors sharply decelerated, two modestly accelerated and only industrials steeply accelerated to a cyclical EPS peak growth rate (Table 5). This EPS breadth deterioration is eerily reminiscent of early-2015 (Chart 7) and is disquieting. Short-term caution is also warranted given the increase in investor complacency. The one sided positioning in the VIX futures market is worrisome. As a reminder, net speculative positions are now at a lower low than the February 2018 level when the VIX snapped to over 50 and caused a massive tremor in the equity market (net speculative positions shown inverted, Chart 8). Table 5Historical/Current/Future Earnings Growth Rates Chart 7Bad Breadth Chart 8Too Complacent But, before getting overly bearish there are some growth green shoots that suggest that Q2-to-Q3 will likely mark the trough in EPS/EBITDA growth and margins (Chart 9). Beyond these positive leading profit indicators, a resolution to the U.S./China trade tussle and China’s trifecta of policy easing measures will also aid in turning profit growth around and really power up U.S. cyclicals’ EPS growth rates. Following up from the January Fed meeting, on February 4 we penned a report titled “Don’t Fight The PBoC” and it is now clear with the recent manufacturing PMI release that China’s easing on all three fronts – credit (Chart 10), monetary (Chart 11) and fiscal (Chart 12) – is starting to pay some dividends. In that light, the U.S. cyclicals vs. U.S. defensives recent outperformance has more room to run. Chart 9Growth Green Shoots Chart 10Chineasing… Chart 11...On All… Chart 12…Fronts   Deep cyclicals have another major advantage this cycle compared with defensives. While at this stage of the business cycle one would expect capital intensive businesses to become debt saddled, cyclicals are still de-levering from the depths of the late-2015/early-2016 manufacturing recession, i.e. paying down debt and increasing cash flow. Defensives, however, are doing the exact opposite with relative cash flow growth problems and piling on debt. Thus, on a relative basis Chart 13 shows that the indebtedness profile clearly favors deep cyclicals vs. defensives. From a bigger picture perspective, while the U.S. has not really purged any debt and it has just shifted it around from the financial and household sectors to the non-financial business and government sectors (Chart 14), the near all-time high in non-financial business sector credit as a share of GDP is disconcerting (top panel, Chart 14). Clearly the excesses are in this segment of U.S. debt and it is unsurprising that debt saddled stocks have been underperforming equities with pristine balance sheets since the 2016 presidential elections (top panel, Chart 15). Such outperformance has staying power, especially given that we are late in the cycle and the Fed has raised interest rates to the point where parts of the yield curve are inverted and a default cycle looms large (bottom panel, Chart 15). Chart 13Cyclicals Have The Upper Hand Chart 14U.S. Debt Profile Breakdown One sub-sector that epitomizes the current cycle’s excesses is commercial real estate (CRE). CRE prices have overshot the historical time trend by almost two standard deviations and it has already been three and a half years since they surpassed the previous all-time high (Chart 16). The recent pullback in the 10-year Treasury yield has pushed cap rates even lower and the bubble in CRE is further inflated. Looking back at the late-1980s pricking of that CRE bubble is instructive and when this cycle ends a big deflationary impulse will likely deal a blow to the CRE market.       Chart 15Hide In Pristine Balance Sheets Chart 16CRE Excesses Are A Yellow Flag Speaking of bubbles, the biggest bubble we currently see is not in equities, but in bonds. Table 6 shows that red is taking over and is reminiscent of mid-year 2016 when the 10-year U.S. Treasury yield troughed a hair above 1.3%. Globally, negative yielding debt is near all-time highs (Chart 17) and the excesses are even larger in the EM sovereign space and in select DM corporates. Mexico raising century debt in U.S. dollars, in cable and in euros is perplexing, as Mexico was at the epicenter of the 1982 LatAm crisis and again in 1994 with the Tequila crisis. Argentina also raising century debt recently in hard currency speaks to the magnitude of the current bond bubble. On the corporate side, Sanofi and LVMH placing negative yielding debt is beyond our understanding, or Total issuing a perpetual bond with a 1.75% coupon. Table 6Red Takes Over   Chart 17Bonds Are In A Bubble All of this is likely linked to the unintended consequences of global QE where fixed income investors are pushed out the risk spectrum and are forced into buying riskier credit. When this bond bubble gets pricked it will end in tears as it always does and the catalyst will likely be the next U.S. recession that will cause a global recession. While our cyclical 9-to-12 month equity market view is constructive and we believe the U.S. will avoid recession, our structural 1-to-3 year view is negative. Nevertheless, we constantly challenge our thesis and the biggest pushback to the negative structural view is the following: What if the Fed can engineer a soft landing in the U.S. as it did twice in the mid-1990s, and the business cycle runs hot for another 5 years (Chart 18)? What if the starting point of low interest rates with the real fed funds rates still close to zero is very stimulative for the U.S. economy as no recession has ever started with a fed funds rate perched near zero (Chart 19)? Finally, what if the late-2015/early-2016 manufacturing recession was actually an economic recession despite the fact that the NBER did not designate it as such and the business cycle got reignited, especially with President Trump’s election that lifted animal spirits? As a reminder, while S&P profits have contracted outside of an economic recession twice before, SPX sales had never achieved that feat, until late-2015/early-2016 (Chart 20). In other words, the revenue recession we had was unprecedented and felt like an economic recession. Chart 18The Fed Has Engineered A Soft Landing Chart 19Stimulative Real Rates Chart 20There Is Always A First Time If that were the case and the cycle were to extend into the 2020s, then the risk is that SPX EPS vault to $200 and valuations overshoot, i.e. the forward P/E multiple spikes to a 20 handle and the SPX catapults to 4,000. In that case, we would leave 1,000 points on the table and our SPX 3,000 view would be way offside. While this is a risk to our negative structural view, there are two sectors we really like for the long-term as we deem them secular growth plays and should do exceptionally well on a 10-year horizon: software and defense stocks. Three key drivers underpin our bullish view on software: galloping higher private and public sector software outlays, a structurally enticing software demand backdrop and ongoing industry M&A (Chart 21). Most importantly, the move to cloud computing and SaaS, the proliferation of AI, machine learning and augmented reality are not fads but enjoy a secular growth profile, and signal that capital outlays on software are in a structural uptrend. With regard to defense stocks, the three key pillars we highlighted in our “Brothers In Arms” Special Report on October 31, 2016 remain intact: the global rearmament is still gaining steam, a space race with manned missions to the moon now includes the U.S., China and India, and cybersecurity is a real threat for governments around the world (Chart 22). On all three fronts, defense stocks stand to benefit as they have beefed up their offerings to provide governments with a one-stop shop solution covering most of these needs. Chart 21Buy The Software Breakout Chart 22Defense Stocks Remain A Long-term Buy     Anastasios Avgeriou, U.S. Equity Strategist anastasios@bcaresearch.com