Marginalia are brief notes written in the margin of a book or article. John Vandenbrink, Deputy Dean of GLOBIS University, will share his notes on topics that have caught his attention in his readings on finance, business, and economics.
Major central banks have created a problem for me in my corporate finance classes. Part of each course involves discounted cash flow valuation, which relies on market rates to determine the value of assets. Today, however, market rates are skewed by central bank policies. In reaction to the global economic malaise, central banks in major countries have pushed interest rates to ultra-low levels. They have altered the normal relationships among the various rates upon which valuations are based, for example, the relationships between nominal and real interest rates and between government debt rates and equity risk premiums. I don't question the need for extreme measures in difficult economic times, but the central banks' actions have shifted the traditional benchmarks and have introduced new uncertainties into investment decisions. When I discuss discounted cash flow valuation in class, I can augment the usual list of qualifications in light of current circumstances, but it must be far more difficult for the many investors and corporate managers who have to discount cash flows to determine what stocks and bonds to buy and what projects to invest in.
Discounting is valuing today cash flows that are expected in the future. The riskier the flows are, and the further in the future they occur, the lower is their value today. The sidebar shows how the math works. It is like compounding interest, but inverted. Arriving at reasonable valuations depends on the validity of both the forecasts and the discount rate. Central banks' policies today are not without impacts on forecasts, but here I will focus on the discount rate.
What determines the discount rate? Conceptually, it is simply the sum of the basic returns that investors demand:
1. Compensation for giving up the opportunity to use their money today (in finance theory, the "real risk-free required return")
2. Compensation for expected inflation
3. Compensation for bearing the risk that some or all of the expected cash flows will not materialize (the "risk premium").
Usually, the levels of the first and second forms of compensation are not separately observable in the market, but together they constitute the nominal risk-free interest rate in the economy, which is readily found by referring to the current yield on short-term government debt, such as US Treasuries or bills. This return is "risk-free" because governments can print money in their own currency and presumably will never default in that currency. The usual reference point is short-term government debt rather than government bonds, but other maturities may be used depending on the particular method of valuation.
The third component of discount rates is the "risk premium" or "equity risk premium." Ultimately, it is simply the return over the risk free rate that the investor judges necessary to bear the risk of owning the asset. However, that alone is not a useful guideline. Usually, the risk premium of an asset is determined by referring to the premiums of other assets of similar risk that are trading in the market. To arrive at the rate for discounting a particular asset, its risk premium is added to the return on government debt, which is the current base level of returns in the economy, and which is presumably more or less independent of the risk premium.
Negative real risk-free returns?
One entry point for understanding the valuation problems that investors (and teachers of corporate finance) face today is the exceptionally low level of nominal risk-free interest rates in major economies. In early September of this year, the yield on ten-year US Treasury bonds fell below 2 percent—the lowest since 1950; the yield on ten-year German Bunds also fell below 2 percent—its lowest level ever; and the yield on UK Gilts fell to the 2.5 percent range—the lowest levels since records began in 1958. At the same time, ten-year Japanese Government Bonds—which had fallen below 2 percent in 1997 and remained there, with only brief exceptions, for fourteen years—yielded 1.0%. Short-term rates were also extremely low. On September 9, 2011, the yield on three-month government debt in the United States was only 0.01%, in Germany 0.57%, in the United Kingdom 0.54%, and in Japan 0.11%.*1 As I write today, October 21, these rates are little changed. Ten-year yields on Treasuries and Bunds have moved to slightly above 2 percent, but all the other rates remain more less the same—and all are very, very low.
As is well known, central banks in major countries have lowered interest rates so far that they no longer have room to stimulate their economies through traditional monetary policies. They have hit the lower bound of zero interest rates. Nor can the governments themselves spend to stimulate, because they face high and growing debt burdens. The response of the central banks has been to inject money into their economies through huge purchases of assets, starting with the debt of their own governments and extending (with differences by country) to mortgage securities, corporate debt, sovereign debt, and even stocks. This practice, known as "quantitative easing," or "QE," has resulted in large increases in the balance sheets of major central banks. The Bank of Japan was the first to embark on quantitative easing. From March 1999 to August 2011, it increased its balance sheet from ¥79 trillion to ¥141 trillion, which is roughly 30 percent of GDP. From March 2008 (before the Lehman shock) to June 2011, the US Federal Reserve increased its balance sheet from $896 billion to $2.87 trillion. That is roughly 20 percent of GDP—only slightly lower than the percentage reached by the European Central Bank in 2011. *2
Ultra-low nominal interest rates coupled with little or no inflation mean that the implied level of real risk-free returns is very low and sometimes negative. This has made it difficult to interpret the real risk-free component of discount rates. What level of real risk-free returns have investors typically required in the past? One answer to this question comes from scholars who have calculated nominal returns and historical inflation over a span of more than one hundred years in major economies. Another answer comes from inflation-indexed bonds, which the UK, US, and some other governments have issued in recent years. These bonds provide a direct measure for observing the real risk-free rate, based on investors' actual expectations for inflation. Both answers suggest that the real risk-free required rate of return tends to range from positive 1 to 3 percent or so. *3 What is the situation today?
In the United States, the real risk-free return—which is, remember, one of the three components of a discount rate—is not just low but negative. This is aberrant. The short-term real risk-free rate, as calculated from the current nominal risk-free rate and the latest reported rate of inflation (September 2011, year on year), is –3.7%. Returns on five-year inflation-indexed Treasury bonds are negative (–0.75%).*4 When the real risk-free rate is negative, instead of being compensated for foregoing use of their money, investors are paying the government for the privilege of lending it their money. In this situation, the government would appear to have an incentive to lend as much as possible: a second money machine for the government, after the printing press. This situation must be unsustainable. Is the market saying that the economic situation is so bleak—that opportunities for productive investment are so scarce—that investors must pay to receive a real risk-free return? Negative real risk-free interest rates can be found in other economies as well. Using again a quick and dirty estimate (current short-term government rate adjusted by the latest monthly year-on-year rate of inflation), real risk-free interest rates are –2.2% in Germany, –0.3% in Japan, and –4.4% in the UK. Is this situation a true reflection of perceptions in the market, or is it a spurious outgrowth of central bank policies bumping against the zero lower bound for nominal interest rates? If it is the latter, then is perhaps some corrective called for when determining discount rates? If so, what? I don't have the answers to these questions.
Distorted risk premiums
Another component of the discount rate that is problematic today is the risk premium. The values assigned as risk premiums depend, we have said, on the riskiness of specific assets being valued. Historically, using data spanning the years 1900–2001, the average premium that investors required to hold a widely diversified portfolio of publicly traded stocks instead of risk-free short-term government bills was 7.5% in the United States, 10.0% in Germany, 9.6% in Japan, and 6.2% in the United Kingdom.*5 One approach is to start from one of these historical total market benchmarks and calculate the risk premium and discount rate for a specific asset by adjusting for the market risk of the asset using the Capital Asset Pricing Model (CAPM). *6 However, not everyone is comfortable starting with a historical total-market value. Another approach is to calculate the implied equity risk premium in the total stock market today and use that as the starting point. *7 Whichever approach is adopted, the value assigned as the asset's risk premium is extremely important in calculating the value of the asset.
Today central banks, through their asset purchasing programs, are intervening in markets on a scale unprecedented in recent times, and in so doing they are distorting prices and risk premiums. This is intentional. They seek to support asset values in order to increase confidence and economic growth. But in terms of valuation, what are the impacts on risk premiums? How can they be measured? Do they require that investors should adjust the inputs to their valuation models?
Economists explaining how quantitative easing works cite two paths that impact risk premiums: a portfolio-rebalancing channel and a liquidity channel. Portfolio rebalancing functions as follows. The central bank purchases large quantities of a specific asset. This reduces the stock of this asset in the private sector and replaces it with cash in the portfolios of investors. This cash, held in banks, is a near riskless short-term asset with low return on investment. In aggregate, the private sector reacts to this situation by investors rebalancing portfolios to restore the previous level of return and risk. This adjustment pushes up the prices of assets across all asset classes. The impact of this rebalancing can be huge. Speaking in March 2010, Spencer Dale, Executive Director and Chief Economist of the Bank of England, gave this assessment of the Bank of England's then recent asset purchases:
A standard portfolio model . . . would suggest that asset purchases on the scale seen might increase asset prices by the order of 20–30%. At first blush, this impact may seem surprisingly large. But it is important to remember that the scale of the asset purchases made over the past year is truly substantial, amounting to some 14% of nominal GDP. *8
Importantly, pushing asset prices up pushes risk premiums down. Paying higher prices for the same cash flows implies a lower discount rate in the market, which, with the nominal risk-free rate unchanged, implies lower calculated risk premiums. Thus, risk premiums derived today from market data are lower than they would be without central bank intervention. Should investors incorporate this information into their valuations? If so, how? If the expected cash flows being valued are to be received further in the future than central bank policies will likely persist, how should investors take this into account?
The liquidity channel of QE also impacts risk premiums, particularly for assets where liquidity is constrained. When central banks enter a market for a particular asset, it is a strong sign of support. Assuring investors that they will have an ability to sell when required increases the value of the asset. At a minimum, central bank support will increase prices and therefore decrease observed risk premiums in the asset class. Liquidity support sometimes borders on price support. Look for example at the Bank of Japan's purchases of stocks. According to the Nihon Keizai Shimbun, the bank buys exchange-traded funds whenever the TOPIX index ends the morning session down one percent or lower. *9 Liquidity provisioning of this sort reduces volatility, shores up equity prices, and lowers observed risk premiums on stocks. Once again, central bank policy impacts a component of the discount rate in a manner that is difficult to measure and problematic for valuation.
The latest twist
The US Federal Reserve recently introduced a new uncertainty for valuations in US dollars. On September 21 it announced a plan, dubbed "Operation Twist," to "purchase, by the end of June 2012, $400 billion of Treasury securities with remaining maturities of 6 years to 30 years and to sell an equal amount of Treasury securities with remaining maturities of 3 years or less." *10 The aim is to lower long-term interest rates, particularly mortgage rates, in order to push up housing prices, and also to improve the climate for long-term investment generally. The result will be lower long-term interest rates and higher short-term interest rates. Practitioners of the discounted cash flow method have differing approaches to their craft, so their valuations will be affected differently by this change. The textbook approach starts from a short-term risk-free rate and adds a risk premium, but others argue for starting with a medium- or long-term risk-free rate to match the cash flows being valued. Both approaches are impacted. What are the practitioners to do now, knowing that a Fed policy—probably temporary—has moved their benchmark?
We are living today in uncertain economic times, under central bank policies that, while perhaps necessary, have shifted the benchmarks that investors and corporate managers have relied on in the past. Even cursory readings about quantitative easing leave the impression that the mechanisms by which QE affects economic decisions—investment decisions above all—are not yet well understood.
*1 Bloomberg.com, 2011-09-09.
*2 Quantitative Easing Issues," M. Miyazaki, Daily Yomiuri, 7 September 2011.
*3 For results based on long-term time series data, see, for example, the results of Dimson, Marsh, and Staunton as reported in Principles of Corporate Finance, Brealey, Myers, and Allen, 9th ed., 2008, p. 174. For results based on inflation indexed risk-free bonds, see, for example, "5-Year Treasury Inflation-Indexed Security, Constant Maturity (FII5)," Economic Research, Federal Reserve Bank of St. Luis, http://research.stlouisfed.org/fred2/series/FII5.
*4 On September 9, 2011, the returns on inflation-indexed Treasury bonds were negative not only at five-year maturity (–0.90%) but even at ten years (–0.03%). Bloomberg.com, September 9 and October 21, 2011. Brad DeLong, who is a professor of economics at the University of California, Berkeley, commented on the implications of negative real interest rates in his blog, "Grasping Reality with both Hands," on September 03, 2011, after the yield on the ten-year inflation protected Treasury bond fell below zero.
*5 "Global Evidence on the Equity Risk Premium," Dimson, Marsh, and Staunton, 2002 (forthcoming in the Journal of Applied Corporate Finance), http://faculty.london.edu/edimson/assets/documents/Jacf1.pdf.
*6 For information on CAPM, refer to a standard corporate finance textbook, or online, see, for example, http://www.investopedia.com/terms/c/capm.asp
*7 "Equity Risk Premiums (ERP): Determinants, Estimation and Implications – The 2011 Edition," A. Damodaran, February 2011.
*8 "QE — one year on," S. Dale, BIS Review, 29/2010. Speech by Mr. Spencer Dale, Executive Director and Chief Economist of the Bank of England, at the Centre for International Macroeconomics and Finance (CIMF) and Money Macro and Finance (MMF) Research Group Conference “New Instruments of Monetary Policy: The Challenges”, Cambridge, 12 March 2010. See also http://www.bankofengland.co.uk/publications/news/2010/027.htm.
*9 The Bank of Japan’s ill-advised “1% rule”, James Saft, Reuters, June 21, 2011.