For a .pdf of this article click here
By Jess Gaspar and Tyler Stevens, Managing Directors, Fixed Income and Commodities, Commonfund
- Investors in commodity programs benefited from the dramatic rise in energy and other commodity prices from 2005 to mid-2008; yet the rapid collapse of credit markets and the swing from inflationary fears to deflationary fears in the second half of 2008brought a reversal of fortunes for commodity investors.
- Since early spring, commodity prices have again surged on demand from Asia and a slowdown in the rate of deterioration in economic conditions in the rest of the world. The debate over commodity market performance in the near term turns on the timing of the switch from concerns about demand weakness in the short run to concerns about insufficient supply in the medium run.
- Adding further stimulus to commodity prices, many worry about significant risks of inflation driven by record monetary and fiscal stimulus; consequently, investors are once again considering allocations to commodities to hedge this inflation risk.
It is hard to fully comprehend recent commodity index returns without looking at the path of inflation. From a peak of 6.5 percent annualized inflation in the second quarter of 2008 to a floor of -12.4 percent annualized inflation during the fourth quarter of 2008.[1] U.S. economy experienced its most intense deflationary episode since the aftermath of WWII.[2] The inflation-hedging benefits of commodities worked suddenly, unexpectedly and perilously in reverse. Ironically, this black period for commodities confirms their critical role as an inflation hedging tool. This letter illustrates these key points and further investigates the relationship between inflation and commodity returns. We hope that it proves helpful as you think about the role of commodities in your portfolio and the timing of commodity investment allocation decisions.
The standard argument for a long-term allocation to commodities proceeds in five parts. First, commodities have historically generated equity-like returns. [3] Second, commodities exhibit low, typically negative, correlations with bonds and equities. Third, commodity index investments are highly liquid investments in highly transparent futures markets. Fourth, commodities provide inflation hedging benefits because they are positively correlated with inflation, whereas nominal fixed income and equities are negatively correlated with inflation. Finally, commodities have historically provided returns significantly in excess of a risk free rate, unlike TIPS which are also liquid and have been mildly correlated with inflation. [4] Note that the first three points comprise a fairly compelling rationale for making an allocation to commodities independent of any inflation-hedging benefits.
The first point is illustrated clearly in Chart 1. Commodity index returns have exceeded both equity and fixed income returns since 1970 and over the last ten years. Furthermore, commodity and equity volatilities are quite similar even if both have higher volatility than fixed income.
Chart 1: Annualized Returns and Volatilities of Asset Classes
For the second point, Chart 2 shows rolling 10-year quarterly correlations of commodities with fixed income and equities. Most of the time commodities exhibit little or negative correlation with these asset classes and the long term average has been about -20 percent. In fact, last year was the only time the rolling correlation to equities has exceeded 25 percent. Even then, the 25 percent level has barely been breached.
Chart 2: 10- Year Rolling Correlations of Commodities with Equities and Bonds
Rolling 20-year and 5-year correlations present basically the same pattern with the same average negative correlation. Each of these correlations exhibits one particularly interesting additional feature. The 20-year rolling correlations had never been positive until the correlation with equities popped marginally above zero in the last six months, confirming the attractiveness of commodities over long horizons. The 5-year rolling correlations reveal that there is one other period which exhibited short term correlations similar to the present: 1983-1984. Remarkably, this period also spanned a sharp drop in inflation rates as the Fed engineered the end of inflation and the arrival of the Great Moderation [5] .
The high correlation between inflation and commodities from 1978-1984 is no coincidence. As Paul Volcker raised rates to rein in inflation, increasingly tight credit cut into demand. The fall in demand caused a fall in commodity price inflation as well, precisely as one would expect. Commodities are expected to perform poorly during demand-driven recessions brought on by tight credit.
We now turn to a discussion of our core topic, the relationship between commodity returns and inflation. Chart 3 portrays inflation since World War I. There are two key points to take home. First, the deflationary episode we had in the second half of last year was worse than anything we have seen since the late 1930s. Second, we have not seen a 4 percent decline in quarterly inflation such as we had during the second half of 2008 since Volcker raised rates to crush inflation in the late 1970s. Even that process took nearly three years, rather than just six months. Unsurprisingly, commodity returns over that period were negative as well.
Chart 3: Three Month CPI Inflation
Now we examine the relationship between commodity returns and inflation. We should expect that commodities are correlated with inflation as they are an element of the inflation indices. Over the last 45 years for which we have data, the correlation between commodities and inflation is 31 percent. To understand the variability about this long term correlation, we plot 10-year rolling correlations in Chart 4. The rolling correlations should be smaller in magnitude because each 10-year period will capture only a subset of the variation recorded in the full history. As expected, the rolling 10-year correlation averages 17 percent and ranges from 60 percent to -20 percent. Five-year and 20-year rolling correlations reveal little additional insight: the average correlations are similar.
Chart 4: 10-Year Rolling Correlation of Commodities with Inflation
We can extend this analysis further by looking at how much protection commodities provide in above average inflationary periods. The long-term median for CPI inflation is 3.7 percent, while the average annualized inflation rate when quarterly inflation exceeds the median is 6.7 percent. During those quarters, commodity returns average 18.6 percent annualized. When quarterly inflation falls below the median, commodities return 3.5 percent annualized, still higher than the average 2.1 percent annualized inflation during these quarters. Thus, commodities provide a higher multiple of inflation in highly inflationary times, significantly adding to their value as a hedge against inflation.
The inflation-hedging benefits of commodities can be seen in other ways as well. Rising inflation tends to be bad for other asset classes, with the correlation between changes in inflation and bond and equities equal to -28 percent and -12 percent, respectively. On the other hand, commodities tend to benefit from rising inflation, exhibiting a correlation of 29 percent. To examine the consistency of the relationship between commodities and changes in inflation, we again plot rolling 10-year correlations in Chart 4, which shows that the correlation between commodity returns and quarter on quarter changes in inflation is more stable than the correlation between commodity returns and the level of quarterly inflation. Finally, since 1970, commodities generate annualized returns of 16.1 percent when inflation is increasing and 6.6 percent when inflation is decreasing.
Chart 5: 10-Year Rolling Correlation of Commodities with Changes in Inflation
This raises the question of whether the level of inflation or changes in inflation is better at explaining commodity returns. To answer this question, we turn to a regression of commodity returns on the level and the change in quarterly inflation. The regression coefficients on these two terms are roughly the same, suggesting that they are of comparable importance in explaining contemporaneous commodity returns.
Thus, the positive correlation between commodities and both inflation and the change in inflation together help explain the poor performance of commodities during the second half of 2008, which is not to say they make it less painful. Restated, commodities revealed the flip side of their positive hedging benefits when the economy slipped into temporary, severe deflation. To be clear, this is how commodities are expected to behave. They are expected to underperform in credit crunches and demand-driven recessions. The trick is to identify when the economy will succumb to these types of pressures in advance.
Inflation is important not just on its own but also in the context of interest rate decisions. In our March 2009 paper, “Asset Allocation: The Case for Diversified Inflation Hedging Strategies”, we highlighted that periods of unanticipated CPI inflation, when real rates of interest are less than zero, are particularly good for commodity returns. In these environments, commodity returns average nearly 20 percent annualized while bonds and equities average roughly zero.
It is not a huge leap to suspect that commodities will also do well when real rates are falling. Empirical analysis confirms this to be historically true. As real rates can be split into two components, the nominal rate and inflation, it may be useful to isolate the behavior of commodities within the two components. We have already shown that rising inflation is good for commodities. It might be tempting to predict that rising short term nominal rates are bad for commodity returns as they slow the economy down. This is not exactly true because investors in commodity futures also earn the three-month Treasury bill return on collateral. Furthermore, declining nominal rates often coincide with an economic slowdown while rising nominal rates coincide with a strong economy. Perhaps counter-intuitively, rising nominal rates have historically been good for commodity returns.
With the information from the prior two sections, one might be tempted to look for a jump in inflation or for the Fed to start raising rates before initiating a long commodities position. Unfortunately, commodities returns lead inflation and not the other way around. Specifically, commodity returns have helped to forecast CPI inflation (and 10-year breakeven inflation) historically. This is not to suggest that one cannot look to inflation for evidence that commodities will likely outperform but rather to warn that such timing may lead to late entry.
Signals from the inflation framework through the first part of this year have been highly auspicious for the performance of the commodity sector. Short term nominal rates are effectively zero. Meanwhile, annualized, seasonally adjusted CPI inflation through the first five months of the year stands at 1.5 percent implying that real rates have been in the vicinity of -1 percent. Furthermore, the transition from deflation last fall to modest inflation this spring implies inflation has been rising and real rates have been falling. Clearly, monetary policy, at least at the front end of the curve, is highly stimulative.
In addition, ten year breakeven rates have risen by 180 basis points this year. While they are still hovering under 2 percent, the upward revision to inflation expectations provides another positive signal about economic recovery. The fact that five-year breakeven inflation rates lie well beneath ten-year breakeven inflation rates provides another promising indicator for commodity returns, this time in the five to ten year timeframe.
Even stolid academics like Ken Rogoff and Greg Mankiw are calling for higher central bank inflation targets for an extended period of time. Some market commentators even argue that the recent experience with deflation in Japan in combination with the U.S. credit crisis imply that unofficial Federal Reserve targets have been set too low and recommend permanent increases in target inflation rates. Their reasoning is that low inflation targets imply a higher probability of hitting the zero lower bound on nominal interest rates during attempts to stimulate a weak economy. [6] Regardless of the reason, higher rates of inflation would erode the real value of debt thus reducing debt service burdens and, implicitly, tax foreign holders to help pay for a U.S. recovery.
As we go forward, the challenge will be how to generate that inflation. Many economists still fear deflation, not least because measures of economic slack including the output gap, unemployment and capacity utilization are loose domestically and globally. As this surplus capacity will not evaporate overnight, short term rates are likely to remain close to zero for quite some time, even with the current expansionary monetary and fiscal policies. This is not to say that the stimulus packages are trivial. The current central bank policy of quantitative easing through printing money to buy private and government debt and the 12 percent U.S. government deficit forecasts for 2009 may have severe inflationary consequences down the road.
Suppose, however, that one sided with the view that headline inflation in the U.S. will remain well contained in the immediate future due to surplus capacity. There are still several important reasons to believe that commodities would do well.
First, high unemployment in the United States, a lack of widespread wage indexing (unlike the 1970s) and a large untapped, low wage labor supply in emerging markets could keep labor costs well contained. Thus, even if raw materials prices experienced substantial price inflation, wages, which represent two-thirds of economic output, need not rise. Hence, the CPI could remain well-behaved even with raw materials price inflation. Second, many observers have been surprised at how quick the recovery in emerging markets has been this spring. As these countries provided most of the incremental demand for commodities over the last ten years and will likely do so for the next ten years, they have capitalized on price weakness to stockpile commodities for future use. Third, spare capacity in commodity producing sectors appears relatively tight compared to other sectors of the economy as has been demonstrated by a variety of market observers. Fourth, capital expenditures in commodity producing sectors have been cut dramatically during the last nine months due to the credit crunch. Future supplies have been hurt by mothballing and permanently shutting down existing capacity as well as by cuts to both capital maintenance and exploration budgets. Fifth, the current shortfall in investment comes on the heels of two decades of underinvestment in commodities.
Finally, it is important to remember the long term commodity picture, specifically demographic trends. It is estimated that the current world population of 6.7 billion will grow by one-third to 9 billion by 2050. This rate of world population growth still understates the magnitude of the natural resource problem as many of the current 6.7 billion are not yet full participants in the global economy.
Billions of developing market workers are entering the global economy without a proportional coincident increase in the supply of raw materials. This shift in relative resource supply will necessitate an upward repricing of raw materials relative to the wage even without population growth. Industrialization, urbanization and simple increased consumption could place strong upward pressure on demand for commodities as these countries grow and prosper. The potential for global instability, while not currently a key investment theme in the markets, provides an additional omnipresent rationale for a long commodity position. The key risks to this view are emerging market stability – can these countries maintain their growth profiles – and technological progress that develops substitutes or discovers unforeseen supplies.
br> In the end, we see significant risks for portfolios based on an asset allocation that would have performed well during the era of relative price stability from 1982-2007 and would encourage investors to acquire protection against both inflation and deflation. In this regard, we believe a well diversified exposure to commodities offers diversification as well as protection from potential inflation threats.
Commodities: Equally Weighted Collateralized Future Index (1970-1990); Dow Jones UBS Commodities Index (1991-2009).
Equities: S&P 500 Index.
Fixed income: Ibbotson Associates Long Term Corporate Bond Index (1970-1972); Lehman Government Credit Index (1973-1975); Barclays Aggregate Bond Index (1976-2009).
Inflation: Bureau of Labor Statistics Consumer Price Index for All Urban Consumers (CPI-U), seasonally adjusted U.S. city average
This report is prepared to inform investors about the views of Commonfund’s advisory affiliate Commonfund Asset Management Company concerning the commodities asset class. The views presented in this report may not be relied upon as a definite statement of trading intent on behalf of any Commonfund fund, or of any Commonfund manager. Forecasts of market or asset class performance do not constitute forecasts of the performance of any Commonfund fund.
Discussion about any investment asset classes or any particular securities should not be relied upon as advice to buy or sell or hold such asset classes or securities or as an offer to sell such securities. This report does not take into account - nor does it provide any tax, legal or investment advice or opinion regarding -- the specific investment objectives or financial situation of any person. The investments, securities, or trading strategies referred to on this report may not be suitable for investors depending on their specific investment objectives and financial situation. Investors should seek professional, financial advice regarding the appropriateness of investing in any securities or investment strategies discussed or recommended on this report and should understand that statements regarding future prospects may not be realized.
Information, opinions, or commentary in this report concerning the financial markets, economic conditions, or other topical subject matter were prepared, written, or created as of the date of this report and may not reflect current, up-to-date, market or economic conditions. Commonfund disclaims any responsibility to update such information, opinions, or commentary.
To the extent views presented forecast market activity, they may be based on many factors in addition to those explicitly stated in this report. Forecasts of experts inevitably differ. Views attributed to third parties are presented to demonstrate the existence of points of view, not as a basis for recommendations or as investment advice. Any market and investment views of third parties presented in this report do not necessarily reflect the views of Commonfund, and Commonfund disclaims any responsibility to present its views on the subjects covered in statements by third parties. Managers who may or may not subscribe to the views expressed in this report make investment decisions for funds maintained by Commonfund or its affiliates.
,br> Investing in commodities entails significant risks. Investors should carefully review the offering materials and other disclosure documents of any Commonfund investment program before investing
The value of and income from your investments may vary because of changes in securities prices or market indices, changes in interest rates or foreign exchange rates, operational or financial conditions of companies or other factors. Past performance is not necessarily a guide to future performance. Estimates of future performance are based on assumptions that may not be realized.
[1] There are two primary ways of calculating quarterly inflation. The standard calculation takes the average CPI during one quarter and divides it by the average CPI in the previous quarter. This variant shows how much inflation has increased on average from one quarter to the next and lines up with economic figures like GDP. The alternative calculation takes the ratio of the CPI at the end of the current quarter to the CPI at the end of the previous quarter. This three-month measure lines up better with asset return data and will be the method employed in this article. In both cases, quarterly inflation rates are annualized by taking them to the fourth power.
[2] In addition, inflation expectations, as measured by U.S. 10-year breakeven rates, also traded in the widest range since the introduction of U.S. inflation-indexed securities in 1997, tumbling from a high of 2.6 percent in July 2008 to a low of 0.1 percent in November 2008.
[3] There can be no assurance that commodities will continue this historical performance. Please see “Important Notes” at the conclusion of this article.
[4] While TIPS have been positively correlated with inflation, one could make the argument that the key correlation should be with unexpected changes in inflation rather than inflation itself. The positive correlation between inflation-indexed bond returns and inflation may not result directly from inflation but indirectly from another factor correlated with inflation such as a liquidity premium.
[5] The Great Moderation refers to the period from 1982-2007 when inflation and long term interest rates fell while the volatility of economic output declined.
[6] To stimulate a weak economy, the Federal Reserve lowers nominal short term interest rates in an attempt to reduce real short term interest rates. This works just fine most of the time. However, the Federal Reserve cannot, using traditional means, lower short term nominal interest rates below zero. This implies that when inflation falls below zero during a downturn, the Federal Reserve is not able to engineer negative real rates to promote economic growth even if it would be optimal. Thus, some economists argue that we should select a higher inflation target, reducing the probability of entering deflationary environments where the Federal Reserve’s short term interest rate policy tool becomes ineffective.
Statements concerning Commonfund Group’s views of possible future outcomes in any investment asset class or market, or of possible future economic developments, are not intended, and should not be construed, as forecasts or predictions of the future investment performance of any Commonfund Group fund. Such statements are also not intended as recommendations by any Commonfund Group entity or employee to the recipient of the presentation. It is Commonfund Group’s policy that investment recommendations to investors must be based on the investment objectives and risk tolerances of each individual investor. All market outlook and similar statements are based upon information reasonably available as of the date of this presentation (unless an earlier date is stated with regard to particular information), and reasonably believed to be accurate by Commonfund Group. Commonfund Group disclaims any responsibility to provide the recipient of this presentation with updated or corrected information.