Ander's Market Theories

[AnderL]

ASSET ALLOCATION FOR BEARS

(Updated May 2004)

We cannot know the future with certainty and this is particularly true when it comes to financial market returns, the economy, interest rates and other matters of concern to investors. We have a large and persuasive body of research showing that markets cannot be consistently timed for profit, that picking stocks or individual securities is unlikely to outperform passive investment strategies, and that eminent economists cannot predict future moves in the economy any more accurately than chance.

A pure passive and index strategy therefore ignores current economic and financial market events and opinions and focuses on probable long-term risks and returns available from highly diversified equity and fixed income portfolios. These risk/return predictions are ultimately based upon past data which invariably contains certain distortions or limitations. Current economic and financial market events are ignored as noise and the focus is on long-term statistical probabilities. But, no matter how thorough academics might be in their analysis of numbers, they too cannot know the future with certainty.

There are a number of economic and financial market factors which arose in the late 1990's and suggest that an exclusive focus on past long-term market return data may disappoint investors putting money into equity markets in the early 2000's. From 1982 through early 2000 investors were rewarded with the biggest stock and bond bull markets in U.S. history. Since we have clear historical evidence showing that big bull markets have been always been followed by big bear markets, I think it wise that passive and index investors at least consider the possibility that we have entered extremely difficult times for equities that may last for a decade or longer, what is commonly termed a secular bear market.

First, there are a few simple ways for passive and index investors to cope with a long-term secular bear market. The last big bear market lasted fifteen years. Investors who received no net after inflation return from 1968 through 1982, and had to endure a 50% decline in equity values at various times during that period, were understandably sceptical that the bear market had ended. How did they fare if they waited until seven years after the bull began, 1989? Quite well; they captured over 80% of the new bull market gains if they were in the market from 1990-2000. There is simply no logical reason to rush new monies into supposed new bull markets. Already existing equity positions should be held since we know we are unlikely to correctly time markets. But, new money allocated to equities can be added gradually over time with little likelihood that most of the gains of a new bull market will be missed. In one sense, investors who are periodically adding money to bear markets are benefitted by them. As the market declines they buy shares at ever cheaper prices. If a new bull market arrives, a bell will not be rung announcing its arrival. Wall Street is often wrong and always optimistic. Taking time to commit money to markets is a form of insurance.

A second simple strategy passive and index equity investors can use if they believe a long-term bear is in place is to maintain a lower equity/fixed ratio than they might ordinarily hold. The overriding determinant of portfolio returns is the ratio of equity to fixed income. Many investors became accustomed to 90% or 100% equity allocations during the late 1990's as the financial services industry incessantly promoted the idea that "stocks for the long run" was always the best way to go. As I point out elsewhere on this site under "Stocks for the Long Run?", there are times when the long run can be far longer than almost any investor has. A 50% equity/50% fixed allocation makes no bets as to whether equities will outperform fixed or fixed will outperform equity, yet captures about 80% per year of what a 100% equity portfolio would capture, with far less risk.

Let's take a look at how a fully diversified DFA strategy, described elsewhere on this site, did during the recent bear market from April 2000 through March 2003. The benchmark S & P 500 lost 40.9%. A DFA portfolio with 40% across all equity asset classes and 60% in short and intermediate fixed income was up 6.84%. A portfolio with 60% in equities was basically flat, no gains or losses. The easiest protection against bear markets is for long-term investors to choose moderate equity allocations and stick with them.

In late 2002 I wrote that we were enmeshed in the worst bear market since the 1930's. It had all the characteristics of a long-term secular bear market. Percentage declines on the major indexes were the worst since the Great Depression. A consistent downtrend since 1999 on the S & P 500 and Dow, and since early 2000 on the Nasdaq had been interrupted by short violent "buy the bad news" and "it's getting better soon" rallies that characterize all bear markets. It is often said that bull markets climb a wall of worry. I would add that bear markets descend a wall of optimism.

As of 2004, the mainstream financial press has declared that the Spring 2000 through March 2003 bear market is behind us. Investors are moving money back into equity mutual funds, speculation in lower rated companies is very strong, IPO's are back, and the economy is supposedly recovering. The Dow and the S & P have risen nearly 25% for the year, with the ever popular Nasdaq up over 50%. Although the future is unknowable as always, there are several problems with this reborn optimism. One is that the bigger the prior bull market, the bigger some of the subsequent bear market rallies are likely to be. Japan had two rallies in the 1990's similar in length and stronger in overall gains than this rally, yet hit new all time lows early in 2003 and isn't much above them at year end. Valuations in U.S. markets are still breathtaking with near record high p/e's and low dividends. Very wealthy and sophisticated investors like Warren Buffet, Sir John Templeton, George Soros and others are betting that the recovery is a phony. So are corporate insiders who run America's corporations; they have been selling stock at all-time record sell to buy ratios. The economic recovery, such as it is, has shown anemic job growth and record breaking debt growth. While GDP growth is being trumpted as 4.5-5% for 2003, debt growth ran over 20%. This is like an individual borrowing $100,000 and proclaiming their income has risen $25,000. This has been a tax-cut and debt driven recovery, a largely jobless recovery, is quite unlike past healthy recoveries, and could easily fail.

If we are still in a bear market, it is likely that this is the result of speculative excess in equities that has not been worked off as of 2004. Excess credit expansion in the 1990's appears to have driven an equity asset bubble which has now spilled over into real estate with record low mortgage rates in 2002 and 2003. When bubbles pop they not only harm asset prices but also damage economies severely. The Fed and U.S. government have followed a highly stimulative policy to prevent a collapse. The usual result of such policies is to delay the inevitable in a cloud of optimism. Optimism is very popular on Wall Street and in Washington. While economists and Wall Street assured investors that all was well with the economy and equities, a promised rebound in the second half of 2001 was replaced by three consequtive quarters of recession. A powerful recovery was again predicted for 2002; it didn't occur. A powerful recovery was predicted for the second half of 2003, and, at best, the decline was halted by eleven Fed rate cuts, big tax cuts, and massive expansion of consumer, corporate, and government debt. The key question is whether the recovery, such as it is, is sustainable or whether a severe bear market in equities will resume.

Several factors suggest that we may be in a long and difficult bear market. A list of bearish concerns, unchanged since 2002, would probably include: valuations that are high to extremely high if conservative accounting standards are used, few signs of a cap ex recovery, global overcapacity in manufacturing and technology, excessive money supply and credit creation, incalculable derivatives risk, heavily indebted U.S. consumers, heavily leveraged U.S. corporations, record breaking increases in U.S. government debt, dependence upon the U.S. consumer as a driver for the global economy, imported deflation, a large and growing current account deficit, a weakening U.S. dollar, a possible debt-driven housing bubble, and adverse long-term demographic trends.

The rest of this article will examine how each major investment asset class might be affected in a severe secular bear market, and how investors might allocate monies to maximize returns in a bearish economy and markets. I've chosen this approach since we have a substantial body of research indicating that asset class allocations are the prime determinant of overall portfolio returns. It seems reasonable to assume that if bulls actively managing portfolios can't, after fees and trading expenses, beat passive and index portfolios, then bears actively managing portfolios probably aren't likely to do so either.

We'll examine equities, bonds, commodities, gold, foreign currencies, home ownership, and hedge funds.

EQUITIES: On first take, it might appear that being long any equities in a bear market is, almost by definition, unwise. The data are very clear, investors starting out at the peak in 1929 received an inflation adjusted return through 1949 of 4.5% per year, including dividends, from the S & P 500. Investors entering the market in 1968 had to wait until 1983, fourteen years, just to break even after inflation. Yet, an investor in the market from 1981 through 2000 received a whopping after inflation return of 12.9% from the S & P. Investing at the beginning of bear markets as opposed to bull markets can make a huge difference in returns, yet a substantial body of research indicates that markets cannot be accurately and profitably timed.

The data above pertain to just one segment of the market, the S & P 500, so let's look at more fully diversified equity portfolios, what many consider to be the most sophisticated passive equity and bond portfolio models available at this time, those of Fama and French and DFA (Dimensional Fund Advisors, Santa Monica). For equities, DFA employs extremely diversified passive equity portfolios targeted at value and small equity premiums in the U.S. and international markets, models derived from factor analyses and multiple regressions on very long-term equity data, in some cases going back to 1926. Domestic and international short-term high grade bonds are included in portfolios to dampen volatility.

Using DFA data, we find that portfolios holding 25% each of US large cap stocks, microcap stocks, one-month T-bills and five year T-notes returned 6.81% from October, 1929, the peak, until ten years later, September, 1938. Deflation averaged -2.25% yearly, so the real return on this portfolio during the worst bear market of last century was 9.05%, substantially higher than the historical average real equity return of about 7%. Volatility was high, of course, with a 23.4% standard deviation. If we look at the inflationary bear market of Jan, 1968 through December, 1981, the results are also surprising. Stocks were widely regarded as having been a loser's investment in the 1970's. Yet, for this period, a portfolio holding 25% each of U.S. large and small, and large and small value stocks, returned 10.24% per year, adjusted for inflation, about 2.7% per year.

More recent data also confirm the value of diversified equity asset class portfolios in bear markets. From April 2000 through March 2003, a 100% DFA diversified equity portfolio was down 14.25% while the S & P was down 40.9%. In 2Q02, the worst quarter for stocks in over three decades, a 100% DFA diversified equity portfolio was down 3.37%, while the S & P 500 lost 13.39%.

These analyses support John Merrill's studies of the Great Depression. He found that a portfolio with 30% in U.S. equities, 50% in bonds, and 20% in cash gave a real return (after deflation in this case) of 7.3% from Sept. of 1929 through February of 1937. Changing the allocations to 47% equities, 47% bonds, and 6% cash yielded about the same real return of 7.4%. This return is very close to the long-term real return on equities from 1929 through 1998, although it should be added that investors had to stay committed through a 4% decline on the first portfolio and 18% decline on the second. It should also be noted that portfolios with 100% in fixed income or 100% in equities did far worse. Thus, even in a very nasty bear market and deflating economy with 25% unemployment, a diversified equity, bond, and cash portfolio gave historically average returns.

Two other equity strategies can be effective in long-term secular bear markets similar to the ones that began in 1929, 1968, and 2000. Investors with equity allocations higher than 35-40% can hedge some or all of their risk by shorting exchange traded funds like the QQQ's, SPY's, and DIA's, highly liquid low cost ETF's which track the Nasdaq 100, S & P500, and Dow Industrials respectively. This will produce a market neutral portfolio or close to it...no gains, no losses, and long-term positions need not be sold. More aggressive or pessimistic investors can simply short these indexes with no offsetting long positions.

BONDS: Fixed income instruments are used to generate income and dampen volatility in portfolios. The historical evidence is very clear. Short-term fixed income securities of high quality, generally running from one to six years in maturity, offer a superior balance of risk and return to longer term and lower quality fixed income securities. From 1802-1997, T-bills gave a real return of 2.9% and long-bonds, 3.5%. Long-bonds, on several occasions, left investors locked into lower interest rates as inflation rose. For example, investors buying long bonds in the 1960's at 4% could only watch helplessly as short-term and money market rates climbed to 14-15% in 1980. As of Spring 2004, Wall Street is expecting a rise in interest rates. It's been wrong countless times before but their could be large losses in long-dated bonds ahead.

In both inflationary and deflationary bear markets and economies, liquidity is highly desirable, and laddered high quality bond portfolios, one to six years out, provide this, while sacrificing little in total returns to longer-term and lower quality fixed income instruments.

COMMODITIES: There is really only one choice for a spot price index which includes oil and gold, and for which we have longer term data, the Goldman Sachs Commodity Index (GSCI). For 1972-1998, the GSCI returned 10.3% per year with a standard deviation (S.D.) of 24.25%. The S & P 500 returned 13.8% per year with a standard deviation of 16.7%, a better return with far less volatility. A 50/50 combination of the S & P and the GSCI produced a 13.4% return with a 12% standard deviation, indicating commodities substantially lowered the volatility of an equity portfolio without substantially reducing its returns. This period included a nasty bear market in the 1970's, inflation, very high interest rates, a weak economy, then robust equity, bond, and real estate markets in most of the period from 1981 through 1998.

The problem for commodities, however, is that a deflating global economy, one bear alternative, decreases demand and commodity prices. In 2001 for example, the GSCI declined 31%, undermining portfolio returns in a year in which almost all major U.S. and global equity indexes were also down substantially, and failing in its role as a portfolio diversifier. However, commodities did do very well in the inflationary economy and equity bear market of 1972-1973, gaining 21.71% and 57.73%, respectively. And, in 2002, commodities were up over 30% in a weak near-deflationary global economy. So, for a bearish investor, commodities might be a diversifier in bear equity markets but their performance is volatile and wholly unpredictable. In 2003, commodities had another strong year and 2004 has been largely up so far, with record commodity prices.

GOLD: A number of bearish writers are fascinated with gold and argue for a return to the gold standard. Very long-term, the price of gold has paced the rate of inflation closely. From 1802-1997, gold returned 1.2% per year and the inflation rate was 1.3% per year, so gold slightly underperformed inflation. T-bills did better, at +2.0% or more after inflation and taxes. For shorter time frames, government interference has produced periods of non-appreciation and rapid speculative appreciation. From 1963-1992, gold returned 7.80% per year, but that includes a very bullish period for gold from 1972-1980, when the U.S. went off the gold standard.

As of Spring 2004, gold stocks and the spot gold price have backed off new highs in the over $400 range after a strong year in 2003. Gold is under half the peak spot price it hit in 1980, around $380 and ounce, but in inflation adusted terms, it's more like a quarter the price. It is entirely possible that it could double or quadruple from its current price, driving gold mining stocks and the metal up dramatically. As a short-term speculative trade, being long gold mining stocks or gold looks appealing, but long-term gold slightly underperforms T-bills, which far more easily protect against inflation. Gold does well in times of rapid inflation and fear.

CURRENCIES: Individual currencies do not really constitute an asset class, but rather the continually varying rates of exchange between various currencies. As of the end of 2003, the U.S. dollar is in a clear and continuing downtrend. This could, of course, poise serious economic problems for America, which depends upon the recycling of dollars back to the U.S. to support its huge trade deficit and consumption of foreign made goods and oil. Although the U.S. is a big net importer of everything, Wall Street has again chosen to take the optimistic path and declare that a declining dollar is good because our corporations will sell more overseas.

From 1968-1992, a period which includes the abandonment of the gold standard and high inflation, the dollar depreciated at an average annual rate of 1.1% per year against the dollar index, a basket of currencies from the G-10 countries and Switzerland. The yen gained 3.58% annually against the dollar for that period, and the Swiss Franc, 3.67%. Neither of these rates of return compare well with returns for stocks, bonds, REIT's, or commodities for this period, so it seems a reasonable conclusion that currencies are not good long term investments in fixed portfolio allocations. However, it's a pretty good bet that the U.S. cannot continue to increase its current account deficit forever, and that the dollar has much more to go on the downside, so as a speculative short-term play and/or "safe haven" for short-term cash, Swiss Francs, backed by gold, and maybe Euros, might be prudent.

HOME OWNERSHIP: Although homes are the major portion of net worth for most Americans, and in some parts of the U.S., returns have been spectacular at times, a home should probably be seen as a consumption, not an investment. That's because it is lived in, and once sold, a replacement will also reflect increased prices, unless one moves to a part of the country where real estate is less expensive or takes cash out. From 1968-1992, the start of Commerce Dept. data, the average one-family house price index appreciated at the rate of 6.17% per year versus 10.44% per year for the S & P 500, trailing both equities and commodities, but outperforming bonds. Sketchier data from earlier in the century suggest that homes and rentals appreciated at about the rate of inflation, and were very sensitive to equity market declines in the 1930's in a deflationary economy. The average California home went from $1,600 in 1900 to $5,600 in 1929, then back to $1,600 in 1932.

In Spring of 2004, real estate is roaring and prices in many parts of the country are up 20% or more year on year. Record low interest rates and a religious belief in ever appreciating real estate have brought home "ownership" to a record number of Americans. Barrons and some housing bears question whether the U.S. has entered into a housing bubble since many investors have begun to view housing as a "can't lose" proposition. Home price increases and affordability are far outpacing increases in income and inflation, which they should track closely over the long run. Further, on a cash-flow/rental basis, real estate is a poor yielding investment in many areas of the country. The CPI inflation data remains benign since housing prices are not included in the cost of living calculations.

Other problems may present themselves when homes are viewed as investments. Should a serious deflationary bear market in equities occur, the evidence is pretty clear that real estate prices decline about as much as stock prices in most circumstances, though they trail the declines in equities. A Bank of International Settlements study constructed asset-priced indexes separate from prices for goods and services, and found that, for 13 industrialized countries, equities and residential real estate generally tracked quite closely. In Japan, for example, Tokyo condos have declined over 70% since their peak in 1990, just about the same as the decline of the Nikkei stock index for the same period. On the other hand, during the inflationary 1970's, equities did poorly and residential real estate boomed in most parts of the US.

Although it's arguable, I don't think one's personal residence should be regarded as an investment, nor as a sure protection of assets if financial markets are collapsing or the economy is deflating.

HEDGE FUNDS: The marketing line for hedge funds is that they outperform all other markets and are minimally correlated with them. Unfortunately, the idea that certain classes of investment can consistently outperform other types of investment is turning out to be a myth, useful for the people selling investments, but dangerous for those buying them. Most data is heavily massaged and there have been a number of notorious hedge fund blowups in recent years.

Research on hedge fund performance doesn't offer an encouraging picture. One study examined venture capital returns from 1960 through 1999 and found that, over 40 years, venture capital funds returned a compound return of 13.4%, one percentage point better than the S & P 500 but one percentage point less than small-cap stocks. Had 2000 and 2001 been included, the figures would have looked far worse. Further, most of these funds have very high fees and survivorship bias in the returns data is a severe problem. One study found failure rates exceeding 75% for a seven year period.

Another study found that the correlation of hedge funds with other markets is far higher than has been generally believed, failing to provide diversification as advertised. Further, hedge funds which make bets on movements of equity, debt, and currencies lagged equity indexes. One study found that hedge funds often hold hard to price illiquid securities, and, after more objective pricing is employed, do not outperform equity markets.

As of Summer 2002, the SEC began an investigation of the largely unregulated hedge fund industry. We heard little about it in 2003 as the SEC had many other problems to investigate. My recommendation is to classify hedge funds as a form of gambling until they are better regulated and reporting standards have been improved, and do not consider them a refuge in a bear market.

What follows is an attempt to integrate the above data into asset allocations for bearish investors. It is not intended to be definitive, nor do I assume that this is the only way to approach bear markets. It is simply intended to provide a starting place for individual decisions. Since it seems to roughly fit the real world continuum of bearish opinions, I'll divide the bearish perspective into three categories, mild, severe, and doomsday.

MILDLY BEARISH: We'll assume that something inflationary like the 1970's in the US occurs; poor economic conditions, poor equity markets, and real estate and/or hard asset classes do well. An allocation with 35% in equities and 65% in short-term fixed would have produced positive though not great returns. A more aggressively bearish mix for these conditions might include 20%-35% highly diversified equities, 20%-50% short-term bonds, 0- 20% commodities, 0-20% real estate, and 0-20% gold or gold stocks.

SEVERELY BEARISH: We'll assume that something deflationary like the 1930's in America or 1990's in Japan is in process. An allocation with 35% in equities and 65% in fixed income would have done well in the deflationary 1930's. A more aggressively bearish mix might include 0-35% long highly diversified equities, 25-50% short-term bonds, 0-25% short equity indexes for short-term speculation, and 25%-40% in foreign currencies and cash, again for speculation and protection from a falling dollar.

THE DOOMSDAY BEAR: We'll assume that something like we've never seen before occurs, a collapse of the global financial system for a period of time, and the beginning of a profound distrust of capitalism. The following asset allocation would then make sense: 25% short-term global bonds including the U.S., 25% short equity indexes for speculation, 25% foreign currencies and cash, preferably the Swiss Franc, the commodity countries, and the Chinese renmimbi, and 25% in gold or gold stocks, all for speculation. Liquidity is king in severe deflations.

I would like to add that I do not endorse allocating assets from any particular market or economic perspective, bullish or bearish, but think considering all alternatives is important in planning portfolios. I undertook writing this paper because several client asked questions about what to do if we're in a serious secular bear market. The major weaknesses of the bullish perspective are its assumption that past asset class returns are sure guides to future returns and that today's equity valuations and macroeconomic conditions don't matter much. The major weaknesses of the bearish perspective are its assumption that specific asset class moves can be predicted in advance and actively managed for profit, and that macroeconomic analysis is a useful tool for enhancing portfolio performance. A further serious weakness is that we have no data allowing us to compare the performance of active bear managers with any passive bear benchmarks. A large body of research suggests that asset class trends and economic variables are unpredictable and cannot be timed, and without a benchmark, results may be due to chance.

Least anyone thinks taking positions on fundamentals, bearish or bullish, is the way to sure profits, consider the Nikkei. It peaked at 38,916 on 12/25/1989. In nine months it was at 20,984, down over 46%. Quick and early players who timed it right and shorted the Nikkei could have done well. But, extensive research suggests that consistent short-term timing of markets is rarely if ever possible. Let's assume our investor held on to their short position, awaiting further profits as the fundamentals worsened, as they did all through the 1990's. The Nikkei made several visits to the low teens only to end up ten years later at 20,434 on 4/10/2000, before it moved under 8000 by March of 2003. As one writer put it, "Its hard to imagine how any of that market action was based on economic fundamentals, because the fundamentals were rotten all the time."

Chase, C. D. Chase Investment Performance Digest. Chase Publishing, 1993

Fischer, D. H. The Great Wave: Price Revolutions and the Rhythm of History. Oxford University Press, 1996

Gibson, R. C. Asset Allocation. McGraw-Hill, 2000

Henderson, D. R. (Ed.) The Fortune Encyclopedia of Economics. Warner Books, 1993

Kindleberger, C. P. Manias, Panics, and Crashes, 3rd Ed.. Wiley, 1996

Merrill, J. Outperforming the Market. McGraw-Hill, 1998



Steven Evanson, Ph.D., R.I.A., Evanson Asset Management

http://www.evansonasset.com/index.cfm/Page/18.htm