Product and Strategy Notes: Depressing Dalbar Study; More 401(k) Fund Choice; Residential Propert: Is a Crash on the Horizon; Poor Man's Alpha; New RYDEX Foreign Currency ETF

A Depressing Dalbar Study

A new study from Dalbar Inc. shows that over the 20 years ended in 2005, the average investor earned an average annual return of only 3.9%, compared to a return of 11.9% on the S&P 500. Why the difference? The average investor had a very bad habit of buying high and selling low, and doing so frequently, which drove up transaction costs. Dalbar makes a critical point: the returns investors realize are as much due to their own behavior as to the performance of the market. And, we would add, the extent to which they have diversified their portfolio. Keep that in mind the next time somebody tries to sell you an investment "that can’t miss!" And then read the next item.

More on 401(k) Fund Choice

In a previous paper, "The Adequacy of Investment Choices Offered by 401(k) Plans", Elton, Gruber and Blake found that many of these plans offered too few asset class choices to their participants. The three authors have just published a new paper examining the quality of plan administrators' fund choices within those asset classes that are offered. The authors find that, "on average [plan administrators] select funds that underperform passive portfolios with the same risk." However, they do manage to "outperform randomly selected funds from the same category." That being said, "plan administrators [also] show less skill in replacing or adding funds. Managers add funds that have performed well in the past and drop funds that have performed poorly...The funds they add are in categories that have performed well in the past relative to other categories (hot sectors). However, after the plans make a change, the preponderance of evidence is that deleted funds did better than added funds, although the differences are not statistically significant."

The authors also analyzed how plan participants allocated their money between the funds they were offered. The authors found that "the sum of [participants'] transfers and contributions was almost exactly equal to the impact of returns in determining the change in participant weights."  This confirms Dalbar’s findings. 

Residential Property: Is a Crash on the Horizon?

Two new studies from respected sources reach a mixed conclusion about this critical question. In "Are House Prices Nearing a Peak?", Paul van den Noord from the OECD concludes his study of multiple housing market as follows: "An increase in interest rates by about one to two percentage points would result in probabilities of a peak nearing 50% or more in the United States, France, Denmark, Ireland, New Zealand, Spain and Sweden." He also notes that, if such a sharp peak is reached, "the historical record suggests that subsequent drops in [housing] prices in real terms could be large and the process could be protracted." On the other hand, he also notes that thus far the U.K. and Australia have managed to cool their respective housing markets without, thus far, triggering a crash.

The second report comes from the Joint Center for Housing Studies at Harvard University. It forecasts a long period of stagnation in the United States housing market, but not a crash. The conclusion is based on the combination of overvaluation (house prices have grown six times faster than median income over the past five years) and high rates along with continued demand for housing (e.g., due to strong immigration) and the fact that about 75 percent of U.S. homes are financed with fixed rate mortgages. Time will tell which view is correct.

"Poor Man’s Alpha"

The IMF's chief economist, Raghuram Rajan, last month offered a fascinating view on the forces driving recent swings in the price of relatively risky assets. At a conference in Spain, he noted that while active managers' compensation is closely tied to their ability to generate alpha (i.e., returns above a comparable index benchmark) most of them found it very difficult to do this. He suggested that this led many of them to pursue the "poor man’s alpha" by simply taking on liquidity risks that others preferred not to hold (e.g., selling deeply out of the money derivatives, or buying low rated tranches of securitized debt structures). The success of this strategy depends on two conditions being met: the absence of low probability, very costly events, and the managers' continued access to cheap funds to finance their leveraged long positions. As Central Banks around the world have raised interest rates, managers' funding costs have risen, threatening their returns (and compensation) and leading some of them to reduce their positions in riskier assets. As these have been sold into relatively illiquid markets (there aren’t many people on the other side of the trade), the result has been big declines in some asset prices. Sounds right to us. And you've got to love that "poor man’s alpha" line.

New Rydex Foreign Currency ETFs

We have frequently written (and again do so this month) about our view that the unwinding of the world's record current account imbalances will likely require both a depreciation of the dollar versus other major currencies, and a prolonged slowdown in U.S. economic growth. As always, our base case view has been that a portfolio that is well diversified across a range of asset classes should be able to perform reasonably well under this scenario, along with many others. In particular, for U.S. dollar based investors, we believe that non-U.S. dollar bonds should help support a portfolio's performance should the dollar weaken versus other major currencies. In our model portfolios, we have used either T.Rowe Price’s (RPIBX) or PIMCO's (PFUIX) unhedged foreign bond funds to implement our allocations to this asset class.

However, as the U.S. dollar has fallen, other products have been introduced, targeting investors who want to profit from this development. Two quite expensive funds from Rydex (RYWBX) and ProFunds (FDPIX) are based on the New York Board of Trade Dollar Index. The underlying currency weights in this index are the Euro (58%), Yen (14%), UK Pound (12%), Canadian Dollar (9%), Swedish Krona (4%) and Swiss Franc (3%). In addition, over a year ago, Everbank launched a range of foreign currency CDs (see our April, 2005 issue). And last month, Rydex expanded its line of foreign currency ETFs, so that they now include not only the Euro (FXE), but also the UK Pound (FXB), Australian Dollar (FXA), Canadian Dollar (FXC), Swiss Franc (SXF), Swedish Kroner (FXS) and Mexican Peso (FXM). These funds have expense ratios of .40%, and earn interest based on the overnight rate in their respective currencies. This rate is less than the rate earned on the foreign bonds owed by RPIBX and PFUIX. Last but not least, Deutsche Bank has registered a long/short currency ETF with the Securities and Exchange Commission that, in theory (and assuming it is ever approved) will function like a currency hedge fund for retail investors (think of it as a Mount Lucas Management MLM Index type product, that only focuses on currencies).

The Rydex ETFs, and the Everbank CDs (which earn a somewhat higher rate of interest, because they lock up your funds for a period of time) create the opportunity for an investor to build their own foreign currency index. With that in mind, we'll offer three perspectives on this issue. The table below shows the respective weights of eight currencies from two different perspectives: a market capitalization weighted index of sovereign bonds, and 2006 GDP weights at purchasing power parity, as estimated by the IMF. Collectively, the eight countries account for about 99% of the total sovereign index’s market capitalization, but only 47% of GDP. In the table below, the weights have been rebased to add up to 100%.

However, bond markets around the world differ widely in terms of the relative importance of sovereign (government) issues. The table below compares the different weighting of government (and government related), corporate and asset backed (e.g., mortgages and other securitized receivables) issues in five Lehman Brothers regional Aggregate Indexes:

When you take all of its segments are taken into account, the U.S. bond market (based on Bank for International Settlements data) accounts for roughly half of the world bond market’s total market capitalization. But that doesn't count bank loans. And as we all know, more and more of them are being securitized today, into collateralized loan obligations and other vehicles. The bottom line is that anybody’s estimate of the "right" division of global bond market capitalization between currency regions is at best a rough guess.

More importantly, it may be irrelevant to the task at hand. As we have described (see "Investing in Debt Markets" ), when it comes to bonds, market capitalization index weighting has some very strong arguments against it (e.g., giving the heaviest weight to the most profligate borrowers). On balance, we prefer two other approaches. The first is used by Goldman Sachs to construct its InvesTop Index. t is an equally weighted collection of bonds that aims to cover a grid of credit ratings and maturities. However, that isn't relevant in the case of trying to construct a foreign index from using either the Rydex foreign currency ETFs or Everbank foreign currency CDs.

Hence we come back to GDP weighting, which, in this context, can be thought of as "relative capacity to generate value", which is then divided between holders of debt and equity issued by the currency zone in question. We admit it's an imperfect approach; however, it at least has some underlying logic to it. If you rebase the GDP weights to exclude the US and still add up to 100%, you end up with the following allocation:

As you can see, these weightings are reasonably close to those used in the New York Board of Trade's Dollar Index. Unfortunately, this allocation presents two problems, one obvious and one subtle. The former is the absence of a Rydex ETF that provides exposure to the Yen. Since Japan's economy has historically had a relatively low correlation with the United States' (and many other countries), this is a serious omission. The more subtle issue is the relatively low weight given to the Swiss Franc. Both GDP and market capitalization weighting underestimate the important role the Swiss Franc plays in the global economy as a low risk asset. As a practical matter, we would therefore give it a higher weight in any index designed to generate high returns when the U.S. dollar declines.

| This Month's Letters to the Editor: Investing in Commodities/Legg Mason and Index Investor's Opinion on Leveraged Index Products | Fundamental and Dividend Weighted Indexes | Product and Strategy Notes: Depressing Dalbar Study; More 401(k) Fund Choice; Residential Propert: Is a Crash on the Horizon; Poor Man's Alpha; New RYDEX Foreign Currency ETF | Economic Warning Indicators Update | Asset Class Valuation Update - Revised - Includes Property, Commodities and Volatility and Updates to Sector Rotation Watch | Global Asset Class Returns | When And How Should You Rebalance? | What Do Bond Spreads Tell Us? | Another Month, Another Crop of New Commodity Products | This Month's Issue: Key Points | Volatility is Up: So What? |