Product and Strategy Notes: New Analyses on Gold as an Asset Class and Financial Advisers' Corner

New Analyses on Gold as an Asset Class

Two pieces of research recently came to our attention that merit another look at gold as an asset class, which we most recently covered in our January 2010 issue. The first is a paper that was written in 2005 by Faugere and Van Erlach, titled "The Price of Gold: A Global Required Yield Theory." The authors begin with a statement that most would agree with: "Assessing the fair value of gold largely remains a mystery in finance." With that in mind, they offer "an asset pricing theory that treats gold as a store of wealth." Their core assumption is what they term "required yield theory" "explains the valuation of financial assets via investors' general requirement to earn a minimum expected after tax real return equal to long-term GDP per capita growth." Based on data from the IMF World Economic Database, we estimate that the average growth rate of real global GDP/capita between 1981 and 2009 was 1.8% per year. We also observe that this very roughly corresponds to the average growth rate for total factor productivity over this period (average annual real GDP growth of 3.3%, less average annual population growth of 1.5%).

The authors further note that, "gold fulfills the unique value of a global store of value...that is, a hedging instrument against inflation and the collapse of value of other asset classes." They then assert that, "(1) the global real price of gold essentially is a real Price/Earnings ratio for gold, where 'earnings' represent purchasing power or a global price index...We define the forward P/E for gold as the nominal price of gold divided by the expected GDP price deflator in the next period; (2) The global real price of gold must vary inversely to all other main financial asset classes real price/earnings ratio in order to preserve the real value of any investor's capital against adverse movements in the values of financial asset classes...Capital flows to gold are dictated by changes in the minimum expected return achievable by other asset classes...Our theory postulates that movements in the global real price of gold occur because of the precautionary demand for gold, which largely depends in the inverse real P/E of other asset classes combined." In practice, however, the authors "assume that the main alternative investment asset class is a stock market index...[Therefore] the global real price of gold must vary inversely with the global stock market forward P/E." Put another way, a change in the earnings yield (E/P) should produce the same change in the price of gold. The authors claim that this "after-tax stock market forward earnings yield [E/P] can be viewed as a minimum expected return...which must equal a required yield given by the sum of the GDP/capita long-term growth rate and the current expected inflation rate." So if either of these changes, so to does E/P and thus the price of gold.

Let's put some numbers on this. At the end of June 2010, the P/E for the FTSE All-World Index was 17.9. Inverting that, the E/P or earnings yield was 5.60%. Subtracting from that our average long-term real growth rate of global GDP/capita of 1.8% leads to an implied expected inflation rate of 3.8%. The just published IMF World Economic Outlook update forecasts an average global inflation rate in 2011 of 2.9%. Adding back our 1.8% long-term real GDP/capita growth rate gives us a target earnings yield of 4.7%. Inverting that gives us a target P/E of 21.3% -- a 19% increase over the current 17.9 P/E. That implies a fall in gold prices of (19%). Applying that to the current gold price of $1,186 gives a forecast 2011 gold price of $961. Of course, if the current fears of deflation are accurate, then the IMF’s forecast for global inflation in 2011 could be much too high. In that case, based on the authors' theory, the gold price would fall by much, much more next year.

So what are we to make of this approach to determining the fundamental value of gold? We have an issue with the author's basic assumption that investors seek the same real earnings yield on all asset classes, which is equal to the long-term average growth of real global GDP/capita (which, as we noted, roughly corresponds to the long-term rate of total factor productivity growth). At the highest level of aggregation, and in the long-run, it is certainly true that total factor productivity growth determines the size of the pie that can be divided between wages paid to labor, returns to capital providers, and taxes. However, even in this case, changes in the bargaining power of these three groups should cause their returns to vary across countries and over time. At a more micro level, in so far as different asset classes have more or less uncertainty and risk, so too should investors require higher or lower real returns to hold them. On balance, we think both theory and evidence weigh against the assumption of "required yield theory."

But that doesn't necessarily mean that gold could not, as the authors assert, serve as a hedge against varying real earnings yields (or, inversely, P/Es) in a given asset class. In their paper, the authors compare gold to equities. In this regard, we noted the following in our January 2010 article on gold's potential long-term role in a portfolio: "In "Is Gold a Hedge or a Safe Haven?", Baur and Lucey distinguish between hedges, where one asset class has a long-term negative or zero correlation of returns with another, and "safe havens" where this relationship only occurs under extreme conditions. Examining U.S., U.K. and German stock and bond prices, they find that "gold is a hedge against stocks, gold is a safe-haven in extreme stock market conditions, but gold is neither a hedge nor a safe haven for bonds." In a subsequent paper ("Is Gold a Safe Haven? International Evidence”), Baur and McDermott find that "gold is a safe haven for major European and U.S. equity markets, but not for Australian, Canadian, Japanese or large emerging equity markets." In light of this, we would say that other evidence only partially supports the authors global contention that changes in equity P/Es  (and inflation) drive changes in real (nominal) gold prices. However, given the outsized impact of the U.S. equity market on global perceptions of uncertainty, we think the author's approach has some merit.

So let's repeat our previous calculations, this time focusing only on the U.S. equity market, where, using the same FTSE All World data, the end of June P/E was 21.9. Inverting that gives an earnings yield of 4.57%. Subtract the 1.8% long-term productivity growth assumption and you get an implied inflation expectation of 2.77%. The most recent IMF World Economic Outlook forecasts 2011 U.S. inflation of only 1.86%. Add the 1.8% productivity growth assumption to this to get an expected 2011 earnings yield of 3.66%. Invert that to get a forecast P/E of 27.32. Since this represents a 24.75% increase in the P/E, it implies, per the authors' theory, a decrease of the same percentage in the price of gold, or a decline to $892/ounce from the current price of $1,186/ounce.

Let's now compare this to another approach to fundamentally valuing gold, which we described in our January 2010 analysis. We will also modify that approach a bit, to incorporate some aspects of the just described "required yield theory." Our approach to asset pricing theory is based on a few key assumptions: (1) Asset prices reflect the interaction of the supply of and demand for real returns from a given asset class; (2) The supply of returns reflects the current yield provided by an asset class, plus expected changes in its price over a given period of time; (3) The demand for returns reflects the prevailing real risk free rate plus a required risk premium; (4) Imbalances between the supply of and demand for returns are normal feature of asset markets; (5) While asset markets are drawn to an equilibrium where the supply of returns equals the demand for returns, they can operate far from equilibrium for extended periods of time; and (6) Asset markets return to equilibrium due to changes in all four underlying variables -- the current yield of the asset, expectations for future price changes, the real risk free interest rate, and required risk premiums.

In our January article, we described we would expect the real price of gold to increase by about 1.75% per year -- the difference between our assumed long-term growth rate of real global GDP of 3.25% per year and our assumed long-term growth rate of the world stock of gold of 1.50% per year. When we looked at the return for holding gold that an investor would logically demand, in terms of a risk premium above the real risk free interest rate, we found that it varied considerably depending on the regime that prevailed. In normal times, the risk premium has been negative (about 2.0% annually), reflecting the fact that gold plays the role of portfolio insurance, for which, in normal times, an investor should logically expect to pay, rather than receive, a risk premium. However, this insurance policy is expected to pay off under the high inflation and high uncertainty regimes, when the risk premium above the real risk free rate turns positive, ranging between 2.5% in the high inflation regime to 2.0% in the high uncertainty regime. Building on the required yield theory, we can further expand our description of the supply of gold returns, viewing 1.75% per year as the normal "income return" from holding gold, and adding to it the change in the price of gold that is driven by changes in perceived uncertainty and expected inflation -- call it the "uncertainty return". We thus have a fully specified supply and demand equation for gold returns, with the return supplied equal to 1.75% plus the uncertainty return under some regimes, and the return demanded equal to the risk free rate plus the required risk premium.

This raises the obvious question of how these variables change to restore the system to equilibrium when supply and demand are out of balance. That is not an easy question to answer. Under the normal regime, the supply demand balance is defined by the difference between 1.75% and the risk free rate less the "insurance premium" investors are willing to pay for gold. If the latter sum is greater than 1.75%, the price of gold should tend to increase. If it is less than 1.75%, the real price of gold should fall. So far, so good -- and, more important, usually quite a stable return generating process. However, when the system shifts out of the normal regime, the gold returns process gets considerably more exciting. On the demand side there is a shift from a negative required risk premium to a positive risk premium, as the portfolio insurance provided by gold is expected to pay off. On the supply side, that should cause prices to rise by more than their long-term normal regime rate of 1.75% per year. The excitement comes when that price increase triggers investor herding, and the price increase exceeds the amount required to match the supply of returns to the demand for returns. As the system is driven further away from equilibrium, with the apparent supply of gold returns exceeding the fundamental demand for gold returns by ever-greater amounts, it becomes more fragile, as maintaining a constant annual percentage increase in price of gold requires ever larger annual dollar increases in the price of gold. Eventually the system is driven back towards equilibrium, with the gold price sharply declining.

We have also noted our view that gold is ultimately a hedge against declining trust in short term U.S. Treasury Bills (and, for some investors, the U.S. Dollar) as the safest and most liquid means of preserving the real value of one's wealth. But consider what happens to the supply/demand equation if that trust is eroded. For the supply of returns, the price of gold is driven up, and perhaps too the associated annual return from holding it. But on the demand side, declining faith in U.S. Treasuries should logically lead to a decline in the risk premium investor require to hold gold even under the high uncertainty or high inflation regimes. In this manner, declining faith in Treasuries only worsens the imbalance between the supply of and demand for returns from holding gold, and causes the gold asset pricing system to become ever more fragile. At the very least, this dynamic suggests that a commitment to systematic portfolio rebalancing is a critical requirement for anyone choosing to use gold as an asset class (as opposed to adding gold coins to the mix of currencies they hold to meet their need for liquidity and precautionary savings, rather than long-term investment needs). Moreover, our analysis also shows that, if one wants to make a long-term allocation to gold as a type of portfolio insurance, the right time to add it to a portfolio is when its price is very cheap, and not when its price has started to rapidly increase.

The second major gold research piece that caught our attention was a special section on the metal ("Store of Value") that appeared in the 10July2010 edition of The Economist. The article notes that "for investors in gold who think of it as an alternative to paper currencies, its attractiveness is intimately linked to their fears about the capacity of these other currencies to retain their value." The article goes on to note that "where the price of gold heads in the future depends on the answers to three questions. First, for how long will investors keep piling into gold? Second, if and when they quit the market, will the demand for gold jewelry revive enough to support the price near recent levels? Third, how will supply respond if the price stays high?" With respect to the first question, The Economist concludes that the answer "lies largely in the sate of the world economy. Western investors' new interest in gold has coincided with the rich world's deepest period of economic turmoil since the 1930s...In a world of unpredictable currencies, driven by fears of massive inflation and with enormous doubts about the true value of many other financial instruments, gold becomes an attractive option."

Moving to the second question, The Economist notes that "at some point, either the worst fears of the gold bugs must be realized -- in which case, heaven help us -- or the world will become a less nervous place. When interest rates eventually rise, the opportunity cost of holding gold of holding gold will go up, taking off the shine...When the overall economic climate improves, so that uncertainty...is no longer so pervasive, that will provide another reason for some investors to retreat from gold." Unfortunately, "traditional markets for gold cannot be expected to pick up the [demand] slack if rich-world investors' appetites should pall." Finally, in response to the third question, The Economist concludes that "if prices [for gold] remain high, more of the world's existing stock will augment [the flow of market] supply. In theory, there is a lot more that could be sold for scrap...[and] the experience of the past year suggests that accounts of India's eternal attachment to gold are somewhat overplayed." In sum, "as long as the world economy remains uncertain and investors fear inflation and sovereign default, gold will keep its allure. Eventually, however, the price will weaken...and investors may look back on the bull run of 2009-2010, or 2009 -2011, with the sort of wonder that humanity has too often reserved for the yellow metal itself."

Advisers' Corner

Every month we review lots of research papers. Most of them are of interest only to other academics. But some of them have practical implications for our readers, either because they suggest new courses of action, or confirm the instincts of the best managers and advisers. The following are brief summaries of research papers that we believe will be of the most interest to our adviser subscribers.

• In "Persistence of Beliefs in an Investment Experiment", Ko and Huang "Test whether investors persist in their positions and inferences in spite of contrary information. They note that "several such biases whereby people accept favorable information and refute unfavorable information, have been documented in cognitive psychology. The first is the confirmation bias, which refers to the tendency to accept information that confirms prior beliefs and refute information that contradicts them...The second is that of the self-attribution bias, whereby people's belief in their own ability persists over time...People attribute success to ability and failure to bad luck...[The third] is motivated reasoning, whereby people passively accept desirable information but actively scrutinize undesirable information in order to contradict it." The authors find that "investors inferences are biased by their prior beliefs in a manner that depends on investment outcomes. Specifically, their perception of new information was more positively biased for their prior favored assets when they were incurring losses rather than gains....Overall, our results indicate that investors engaged in motivated reasoning to justify their prior choice of favored stock-- i.e., they actively distorted new information about this stock more for losses than for gains...[This type of motivated reasoning] may also contribute to slow incorporation of new information in the price of loser stocks...Our results indicated that investors' beliefs should be more persistent when losing money, so that momentum ought to be stronger for loser than winner stocks. [Consistent with this], several papers have documented the fact that momentum is driven primarily by persistence in losers."
  
  
• De, Gondhi, Mnagla and Pochirajo cover similar ground in "Success/Failure of Past Trades and Trading Behavior of Investors". They find that "the success or failure of a trade, where a profitable trade indicates success and an unprofitable trade failure, influences the current trading decisions of investors [both the frequency and size of subsequent trades] more than the size of the gains or losses from the trades." They also find that "on average, this bias reduces the profits from current trades for investors." Similarly, in "Overconfidence Among Professional Investors", Putz and Ruenzi "find that mutual fund managers trade more after good past performance. This behavior is driven by good individual portfolio performance, while the market performance has no significant impact."
  
  
• In "Security, Potential, Goal Achievement and Risky Choice Behavior", Isacco Piccioni proposes a new models for risky choice behavior that captures some of the key assumptions that have long guided our thinking about asset allocation and risk management. More specifically, Piccioni "emphasizes agents' twin desire for downside protection and upside potential, as well as the importance of goal achievement in risky choice behavior. More technically, Target Utility Theory "is characterized by three main features. First, the utility function is defined over gains and losses with respect to a reference point or target returns. Second, investors are driven by the twin desire for downside protection and upside potential. Third, individuals have a greater preference for those choices that allow them to avoid more regret."
  
  
• In "How Investors Face Financial Risk: Loss Aversion and Wealth Allocation", Rengifo and Trifan "focus on the decision process of non-professional investors." They reach a number of interesting conclusions that will resonate with many professional advisors. These include: (1)"Past performance drives the current perception of the risky portfolio." (2) "The main source of fluctuations in investors perceptions of the risky portfolio is the frequency at which its performance is evaluated." (3) "The proportion of risky investments in the portfolio decreases quickly when investors check portfolio performance more than once per year." (4) Evaluating portfolio performance once a year "is optimal for generating positive attitudes towards risky investments." And (5), "Many studies might underestimate non-professional investors’ actual aversion to financial losses."
  
  
• Does working with smart people improve your own performance? In "Peer Effects on Analyst Forecast Accuracy", Constantinos Antoniou of Durham Business School "examines whether security analysts' forecast accuracy is positively related to the accuracy of their colleagues." He finds that the effect is positive and significant, and "becomes more magnified when the forecasting task becomes more difficult."
  
  
• What drives rebalancing behavior in practice? In "The Impact of Risk and Return Perceptions on the Portfolio Reallocation Decisions of Mutual Fund Investors", Ederington and Golubeva begin by noting that, "a series of studies have found that future stock market returns and/or volatility are partially predictable from financial and macroeconomic variables, such as the current dividend yield or interest rate term structure...[Other studies] have found that the optimal portfolio allocation [between cash, bonds, and equities] of hypothetical investors should also be sensitive to changes in these variables." In the current study, the authors test these predictions against the behavior of real investors, specifically the way they reallocate assets between money market, bond and equity mutual funds. The authors find that "reallocation between equity and bond funds on one hand, and money market funds on the other are a strong positive function of the slope of the interest rate term structure." However, they "find little reaction to changes in the dividend yield." Rather, "in the aggregate, mutual fund investors tend to act as momentum investors as regards both stock and bond markets. When stock markets are high (negative) there is a strong net flow into (out of) equity funds from (into) money market funds. When bond market returns are high (negative), funds flow into (out of) bond funds from (into) money market funds." However, "equity and bond funds returns have only a small impact on flows between equity and bond funds" and "when long term interest rates rise so that bond prices fall and bond fund returns turn negative, funds tend to flow out of bond funds despite the rise in yields-to-maturity, underscoring the momentum behavior of bond fund investors." Also, "net equity fund exchanges are negatively correlated with changes in the VIX, and net bond fund exchanges are positively correlated...with a smaller impact on net exchanges with money market funds." Finally, the authors "find no evidence that mutual fund investors rebalance to restore their original portfolio allocation percentages following large stock or bond market movements. In other words, if a sharp stock market rise causes the actual equity percentage in an investors' portfolios to rise, they move funds into equity funds, not out."
  
  
• Here's some interesting research on mutual fund advertising: In "Worthless Warnings? Testing the Effectiveness of Disclaimers in Mutual Fund Advertisements", Mercer, Palmiter, and Taha observe that "mutual fund investors flock to funds with high past returns, despite there being little, if any, relationship between high past returns and high future returns. Because mutual fund management fees are based on the amount of assets invested in their funds, however, fund companies regularly advertise the returns of their high performing funds." Because of that, "The Securities and Exchange Commission requires fund advertisements to contain a disclaimer warning that past returns don't guaranteed future returns and that investors could lose money in these funds." After reviewing the available evidence and conducting further experiments, the authors conclude that the current SEC mandate "is completely ineffective. It neither reduces investors' propensity to invest in advertised funds nor diminishes their expectations regarding the funds' future returns." The good news is that the authors also conclude that "a stronger disclaimer -- one that informs investors that high fund returns generally don't persist -- would be much more effective."
  
  
• Sometimes we come across studies that leave us shaking our heads in amazement. Here are two recent ones. "In Physical Contact and Financial Risk Taking", Levav and Argo find that "minimal physical contact can increase people's sense of security and consequently lead them to increased financial risk taking behavior." They also found that this effect is particularly strong when the toucher is female, and weakens when a handshake is used instead of a subtler touch (a pat on the shoulder). As Argo noted in a separate interview about the study, "a woman's touch inspired feelings of comfort, confidence and security in the participants...It's a gentle touch, whereas a handshake is much more businesslike and formal -- it signals something different...You might not actually stand there and say, 'Boy, that felt like what my mom used to do.' It happens instantaneously at an unconscious level and, as a result, makers you feel more secure so you're willing to take risks."
  
  
• In "A Corporate Beauty Contest", Braham, Harvey and Puri had "close to 2,000 subjects study the facial traits of CEOs." They found that "subjects rated CEO faces as appearing more competent and less likeable than non-CEO faces...Another experiment matched CEOs from large firms against CEOs from smaller firms and found that large-firm CEOs looked more competent and likeable...A third study found that executive compensation is linked to perceived competence ratings." The authors went on to explore the underlying factors driving facial-trait rating, and found that perceived "maturity" and "baby-facedness" were important drivers, "with the more mature faces assigned higher competence ratings." Finally, after examining corporate performance data, the authors "could find no evidence that the firms of competent-looking CEOs performed better. Essentially, the 'look' of competence says very little about effective competence." In a similar study ("Stocks of Admired Companies and Spurned Ones"), Anginer and Statman "studied Fortune magazine's annual list of 'America's Most Admired Companies' and found that stocks of admired companies had lower returns, on average, than stocks of spurned companies from April 1983 through December 2007." They also found that "increases in admiration were followed by lower returns." Finally, among the spurned companies, "the dispersion of returns was high...implying that investors who want to benefit from the return advantage of spurned companies must diversify widely among them."
  
  
• Here's a good study to keep in mind the next time you wonder what is really going on inside your client's head. In "What You Don't Know Won't Hurt You: A Laboratory Analysis of Betrayal Aversion", Aimone and Houser find that the impact of betrayal aversion is substantial. "Holding fixed the probability of betrayal, the possibility of knowing that one has been betrayed reduced investment in a risky asset by about one-third." This helps to explain why betrayal aversion "leads many people to avoid risk more when a person, rather than nature, determines the outcome of uncertainty." This may also help to explain the psychological attraction of ETFs rather than individual stocks recommended by an advisor.
  
  
• The previous study is also interesting in the context of a new study done in Canada, where BlackRock recently released interesting findings about the views of high net worth investors. "More than 80 percent of HNW respondents felt it was important that financial planners put their client's interests first...76 percent of HNW investors said they turned to advisors for at least some advice...64 percent said they were reevaluating their portfolio mix...70 percent of HNW investors who are familiar with ETFs say they provide a significantly better rate of return than mutual funds...However, only 27 percent say their advisor, broker or financial planner has recommended that they buy an ETF...Nearly half of the 48 percent of HNW investors who said they own mutual funds believe that their mutual funds did not charge fees in the form of management expense ratios [!!!]...63 percent of HNW investors under age 35 felt it wasn't worth paying advisors for advice, compared to just 26 percent of those over 35."
  
  
• Another study may point to how advisors can improve their clients' satisfaction. In "To Do or to Have? That is the Question", Van Boven and Gilovich asked whether experiences make people happier than material possessions. They conclude that this is the case, and that "experiential purchases -- those made with the primary intention of acquiring a life experience -- made people happier than material purchases...[Because experiential purchases] are more open to positive reinterpretation, are a more meaningful part of one's identity, and contribute more to successful social relationships." When we read this, we thought of our observation that medical doctors seem to divide into two groups when confronted with patients who arrive for a consultation armed with a substantial amount of research they have collected online. One group, which seems more science oriented, paternalistic, and resentful of "patients trying to practice medicine." The other seems more oriented to engaging the patient in a relationship focused on improving health, and is much more encouraging and enthusiastic about online research that contributes to this outcome. In our experience with both types of doctors, we have that the former is more likely to be older, male and located on the East Coast of the U.S., while the latter is more likely to be younger, female and located on the West Coast. Overly broad generalizations to be sure, but still suggestive of the concepts that underlie this article. The relevance of this research to financial advisors seeking to grow their business seems clear -- that offering clients a value proposition based on "doing" (e.g., learning to become an astute investor) is likely to be a superior value proposition to one that is focused on "having" (e.g., my advice will result in higher returns).
  
  
• Finally, many advisers know that one of the most challenging aspects of their job is to help clients keep money in the right perspective. This study should help you to do just that. In "Money Giveth, Money Taketh Away", Quoidbac, Dunn, Petrides and Mikolajczak "provide the first evidence that money impairs people's ability to savor everyday positive emotions and experiences. In a sample of working adults, wealthier individuals reported lower savoring ability (the ability to enhance and prolong positive emotional experience). Moreover, the negative impact of wealth in individuals' ability to savor undermined the positive effects of money on their happiness." In fact, when the authors "experimentally exposed participants in their experiments to a reminder of wealth", it "produced the same deleterious effect on their ability to savor as that produced by actual differences in wealth." The authors conclude that their evidence "supports the widely held but previously untested belief that having access to the best things in life may actually undercut people's ability to reap enjoyment from life's small pleasures."

| Table: Fundamental Asset Class Valuation and Recent Return Momentum | July 2010 Issue: Key Points | This Month's Letters to the Editor: Why the mix of .70 Plum Creek Timber and .30 Rayonier to Implement Your Allocation to Timber, instead of CUT ETF; How Do I Use Your Global Asset Class Valuation Analysis?; and Could you please clarify what you mean in your monthly Equity Valuation Analysis by "Low Demanded Return" and "High Demanded Return"? | Feature Article: Understanding and Predicting Uncertainty Shocks, Part 2 | July 2010 Economic Update | Investor Herding Risk Analysis | Global Asset Class Valuation Updates Detail through June 30, 2010 | Product and Strategy Notes: New Analyses on Gold as an Asset Class and Financial Advisers' Corner | Overview of Our Valuation Methodology | Uncorrelated Alpha Strategies Detail | Global Asset Class Returns | Table: Market Implied Regime Expectations and Three Year Return Forecast |