How the Economic Machine Works
( Ideas expressed in this section and the title are from title of Ray Dalio’s Economic Paper available at, http://www.economicprinciples.org)
We recommend Ray Dalio’s “Economic Principles” document, it is a highly informative read. The document offers insight on how the worlds largest hedge fund makes investment decisions and views the global economy. We hate to summarize the document as doing so removes many of the nuances contained therein, however the document describes the relationship between debt and economic growth superbly, therefore a very brief summary of one of its sections is an excellent introduction to this edition of Quantessential Views.
Debt is key to Mr. Dalio's understanding of the economy. In his view there are two main types of economic cycles: a short term debt cycle, and a long term debt cycle. The short term debt cycle is the typical 5-7 year cycle ending in a recession triggered by a tightening in private sector credit expansion through a tightening in monetary policy. The tightening is necessary because the top of this cycle is characterized by demand exceeding capacity to produce, increasing inflation beyond a desired range. This is the business cycle most of us are familiar with. The long term debt cycle occurs every 50-75 years and ends when debts rise at a much higher rate than incomes and can no longer be serviced.
Mr. Dalio believes that we began to experience the end of a long term debt cycle in the 2008/2009 recession. The last time this occurred in the USwas in the 1930s. In this edition of Quantessential Views, we look at the significance of debt, briefly look at the 1930s and highlight parallels to today, surmise what may come next, and discuss how we are positioning our portfolio to take advantage.
Economic correlations and the importance of debt
Figure 1 shows correlations between various economic statistics and asset prices between 1945 and 2000. In this timeframe, the ability of the economic system to take on more debt and service existing debt was high. For most of this period, debt was growing at the same rate as income, and in periods where it was not it was possible to lower interest rates to ease debt servicing pains and allow more debt creation. As can be seen, in this period, debt levels are highly correlated to SP500 returns, GDP, and CPI.
Figure 2 shows correlations between the same economic statistics and asset prices between 2000 and 2008, a period where debt levels in the economic system were high. To refer to Mr. Dalio’s document, this period is the tail end of the long term debt cycle: debt creation is escalating at a faster rate than incomes, and interest rates are near zero and generally heading down. In this timeframe, debt levels and equity values were less correlated. Interestingly, in this period a relationship between VIX and junk bond yields develops.
Figure 3 shows the same correlations between 2008 and 2Q16, a period where some deleveraging occurred, but mostly debt levels changed at the same rate as incomes. Historical correlations between debt and equity returns, GDP and CPI return; the relationship between VIX and junk bond yields holds.
Figure 1: Correlations, 1945-2000
Source: Fred, Yahoo, Vital Data Science Inc.
Figure 2: Correlations, 2001-2008
Source: Fred, Yahoo, Vital Data Science Inc.
Figure 3: Correlations, 2009-present
Source: Fred, Yahoo, Vital Data Science Inc.
The purpose of these correlations is to highlight that at different debt levels asset returns are influenced by different macro economic statistics. However, in general, debt is a good proxy for economic growth and equity returns.
Lessons from the 1930s
After the 1929 market crash, the subsequent depressionairy period of the early 1930s was characterized by deflation, bank runs, and debt crisis. Authorities reaction to the crisis was limited by the dollar’s link to gold, and deflationary policies like increasing taxes and lowering spending. In 1933 Roosevelt took office, effectively removed the dollar off the gold standard, and instituted a number of stimulus programs - including printing money and giving it directly to citizens through initiatives such as the Home Owners Refinancing Act and the Farm Credit Act. The economy responded - asset and prices increased, GDP growth accelerated, unemployment fell, and inflation started to increase.
However, in 1937 a number of events occurred which caused a economic contraction. These events are summarized by Roose in Economics of Recession, reprinted in A Monetary History of the United States, 1867-1960, Friedman, Schwartz:
“In broad outline, the causation may be reduced to a relatively few important elements… [N]et government contribution to income was drastically reduced in January 1937…
[A]t the same time… the Federal Reserve action on excess reserves caused short-term governments to weaken and set up thereby a chain of reaction which resulted in increased costs of capital and the weakening of the securities markets to which business expectations are very sensitive, especially in the United States. The operation of the undistributed profits tax, in addition o its effects on business expectations, also reduced the cash position of even the large companies. The imperfect supply of capital funds and their increased cost made it more difficult for borrowers to obtain capital. Most important of all, however, was the reduced profitability of investment, beginning the first quarter of 1937. This resulted from the increased costs, in which labor costs played a prominent part…”
Friedman & Schwartz, A Monetary History of the United States, 1867-1960, pg. 543
When the 1937 contraction hit, the Federal Reserve System lowered the reserve ratio again and continued to print money. Come 1938, the Federal Reserve Systemwas out of traditional tools to stimulate the economy as reserve requirements were low and the discount rate was near zero. However, at this point World War II broke out in Europe and the US economy was first reflated by demand from European allies, then by federal government demand when the US joined the war. The post war years saw the global economy benefit from many factors including favourable demographics, the rebuilding of Europe, and the globalization of supply chains.
What makes the 1937 recession interesting is parallels to the current period. Similar to today, the US economy recently went through period of deleveraging, economic stimulus has been employed liberally and was reaching its limits, there were expectations of inflation in the near future, tightening monetary policy, and a fragile global economy.
The key lessons from the 1930s which we believe are relevant to today’s period are:
- Deleveraging takes time (>10 years); a deleveraging may start acutely when an economic bubble bursts, however it is an economic event which will take place largely in the ‘background’ of economies for a long period;
- Short term asset and economic cycles will occur during deleveraging;
- Many of the tools used to ease the pain of deleveraging lower currency values, when many nations employee these simultaneously, currency and trade battles have a greater chance ofoccurring;
- Economies are highly sensitive to tightening monetary policy movements during a deleveraging; and,
- Governments have fiscal stimulus tools at their disposal, which can be generalized in two categories: a) creating demand (i.e. defence and infrastructure spending); and b) giving money to citizens (i.e. providing low interest loans or erasing loans); these can be employed in conjunction with or irrespective of monetary policy.
Back to the future
Chart 1 shows US household, business, and government debts to compensation, profit, and receipt ratios, respectively. These values were calculated using each groups corresponding GDP contribution. Chart 1 represents the relevant debt/GDP ratio for households, business, and government. The purpose of creating the chart is to highlight: 1) business and household debt levels relative to income have come down since the ’08/’09 recession, however remain near historical highs; 2) Household/Business debt reduction has been offset by government debt increases; and 3) Government debt to income has stabilized, while household and business could be showing signs of increasing.
Chart 1: Household debt/compensation, business debt/profit, gov’t debt to receipts
Source: FRED, Vital Data Science Inc.
Chart 2 provides a specific view of the combined data: we graph the cumulative difference between debt/GDP growth percentages over the time period and overlay the Fed Funds rate. The purpose is to highlight periods (shown as a negative slope) where debt growth exceeds GDP growth. Highlights include: 1) over the chart’s time frame, debt has generally either grown inline with GDP or at a higher rate; 2) since 2008 total debt has generally grown inline with GDP, however some deleveraging took place in the 2010-2012 timeframe; 3) interest rates can no longer be used to stimulated debt growth.
Chart 2: Cumulative % delta between debt and GDP, Fed Funds rate
Source: FRED, Vital Data Science Inc.
Prior to the election of Trump our belief was that the slow deleveraging which started after the 2008 recession would continue, along with slow economic growth, job creation, and income growth in the short to medium term. Based on this, our belief was that equity prices had run ahead of themselves. Our current view is Mr. Trump’s administration picks, expected financial deregulation, potential fiscal stimulus, and his belief that banks are not lending, combined with relatively low interest rates and increased ability of consumers and business to lever up may create an environment in which debt creation will ramp up. This would fuel consumption, inflation, and corporate earnings in the short to medium term. Given that the Fed has been supportive of economic growth and is not fearing inflation we believe that this phase of high debt creation can last for a multi year period.
Chart 3 shows the delta between potential and actual GDP, recessions, and SP500 returns. The chart highlights deflationary and inflationary gaps, which we believe are a good predictor of monetary policy and recessions. An inflationary gap exists when actual GDP > potential GDP. This causes inflation outside of the Fed’s target, the Fed raises interest rates, debt growth is slowed, demand drops, markets correct, a deflationary gap is created, interest rates are reduced, the cycle continues. By this measure (currently a deflationary gap) the US economy has room to heat up before inflation becomes a problem and monetary tightening becomes aggressive.
When they occur, inflationary gaps generally last slightly over 2 years; however historical data exhibits a large standard deviation (over 1.5 years) with a right skew. Therefore, when an inflationary gap opens, the balance of probabilities is that condition exists for a period that is shorter than the average of ~2 years.
Chart 3: Inflationary and deflationary gpas
Source: FRED, Yahoo, Vital Data Science Inc.
Both household and businesses have the ability take on more debt as both reduced debt levels since the financial crisis, however, debt to income levels remain near peaks. Further, currently earnings and income growth is relatively tepid, and regardless has fallen behind debt loads in the past several decades. So while households and businesses have the ability to take on more debt and spend it, debt fuelled spending/investment will likely be short lived.
Europe is deleveraging. The deleveraging in Europe started later than in the US (chart 4), and was more complicated due to the nature of the economic union and the decision to not adopt quantitative easing and fiscal stimulus early after the financial crisis. As a result, Europe is behind the US in macro economic terms: inflation is low, job creation is minimal, unemployment is high, and GDP growth ranges from negative to anemic in most countries.
China is behind further in their debt cycle as it continues to lever up (chart 4), and we believe that despite the current blasé attitude, events in China have the potential to be a negative catalyst for global macro conditions in 2017. Debt rates relative GDP are high and growing (charts 4); currency is flowing out of the country putting pressure on yuan (chart 5); and trade wars are brewing. Recent denial by the EU and US of ‘free market status’ in the WTO, combined with the potential of the new US administration to label China a currency manipulator, and China’s recent tweaking of the basket of currencies it pegs the yuan (lower the contribution of the USD) have the potential to increase trade tensions between the US, China, and the EU.
Chart 4: Domestic credit to private sector as %GDP
Source: World Bank
Any of the following Chinese scenarios, or a combination of them, would cause investors to increase risk premiums with spillover effects to the global economy: 1) debt bubble bursts causing tightening credit conditions; 2) capital outflows put pressure on currency causing Chinese companies with US debt to having issues servicing the debt and tightening credit conditions; and 3) trade provocation with the US slows trade, growth, and adds to global uncertainty. In our view, these are probable near/medium term events which are not currently priced into asset valuations.
Chart 5: Capital outflows from China
To take of advantage of our economic outlook, we have positioned our portfolio to be long US equities, hedged with gold; and short positions in CDS.
Globally, we believe that many of the tools which governments and federal reserve banks will continue to employ will put downward pressure on currency values. Further, with equities trading near extreme valuations, and inflation and rising interest rates putting pressure on fixed income, through the process of elimination we see gold coming into favour with investors in the near to medium term. Given this view and golds historical low correlation to equities we believe it is a reasonable hedge.
We have further hedged our equity positions by taking the short side of CDS positions. We like the risk/reward of this trade because the cost is lower than purchasing long term puts or shorting (without entering complex cost offsetting arrangements); the high correlation between volatility and junk bond debt yields; and, since the position is likely to have some intrinsic value under most scenarios, we are comfortable holding the position for a longer period, reducing our dependence on timing the markets.
Any material provided in this note is for general information use only. You should not act based solely upon the materials provided herein. Vital Data Science Inc. advises you to obtain professional advise before making investment decisions. Your use of these materials is entirely at your own risk. In no event shall Vital Data Science Inc, its officers, directors or employees be liable for any loss, costs or damages whatsoever.
Subscribe to receive further research
Unsubscribe to be taken off of our list