The other day I was trying to develop a mid term strategic allocation for new money when I began to develop an idea about the current investor risk appetite. Being somewhat bearish on equity markets in the mid term, I came to a difficult decision in how to allocate to equities, fixed income, and cash. The most common assumption in most economic models is that "rational investors are risk averse" - meaning they prefer less risk to more risk in order to get a given amount of return. Our reliance on such assumptions in building the current market we live in may have caused a phenomena which I call "risk perversion" and "risk inversion".
Risk Perversion is a perverse view of the risk of an investment. Jeremy Siegel in his book "Stocks for the Long Run", marking to model, and the current trend towards Asset Allocation by Financial Advisors are one of the main driving forces in risk perversion. In Stocks for the Long Run, Jeremy Seigel made a case that long-term oriented equity investors should expect an average annualized return of 7% a year. He elaborates on the fact that as long as an investor keeps his time horizon in perspective, equities are the best risk/reward combination available to the average investor. He writes,
In 1991] began to research the returns on financial assets to determine whether stocks did realize superior returns for the long term investor. My examination into nearly two centuries of financial data reveals [that/ although stocks are certainly riskier than bonds in the short run, over the long run the returns on stocks are so stable that stocks are actually safer than either government bonds or Treasury bills.The constancy of the long-term, after-inflation returns on stocks was truly astounding, while the returns on fixed-income assets posed higher risks for the long term investor. This is because bondholders can never be compensated for unexpected inflation, a factor that cannot be ignored in our world of paper money.
This is all good in fine in isolation however, the theory of an Equity Risk Premium is also very prevalent in modern financial theory. The perversion of risk therefore surfaces in that in one case investors are expecting a premium in their investment returns over the risk-free rate by investing in stocks, and in another case they expect that stocks are essentially the risk-free rate. This is complicated by perversions of inflation data as mentioned by Barry Ritholtz today on TBP. When the risk free asset becomes what is truly a risky asset, the idea of risk is perverted to a great degree. This creates a self-reinforcing scenario where true equity risk premium slowly erodes, and inflation must increase in order for investors to realize their "long term expectations". This, known as the Equity Premium Puzzle causes investors to find false solace in the fact that they have met their return expectations, shielding them from the idea of the risk they are actually taking. Sound Familiar? An excellent illustration of this perversion of risk is shown on a chart published by CXO Advisory Group back in January. I have posted it below
Notice how the VIX, the investor fear gauge, slowly falls along with the ERP. Not present in the chart however most people can draw the line in their mind is the line for inflation and dollar devaluation along the same time period.
In addition to this, the widespread acceptance of "marking to model" confounds this relationship further. Marking to model is widely known through the whole CDO blowup a few months prior where investments got their intrinsic value from a model rather than a market. In this context, I will extend that further to the idea of Asset Allocation permeating all the Wealth Management Dens at the moment to Expected Risk and Return.
After the market blowup in the early part of this decade, Stock Brokers became Wealth Managers, and instead of pitching stocks they now pitch Asset Allocations. Biased stock research has now been replaced by biased backtesting of Asset Allocations coupled with Financial Plans - all of which are largely sales tools. Speaking from experience, the asset allocations given from one large brokerage house gives returns for the last 20 years. Unless you have been living under a rock for the last 20 years, you would know that it has been marked by an unprecedented of rally in both stocks and bonds, and having exposure to both has been a winning scenario which has been fueled mostly by a Fed that was very astute at running the printing press. Now we have many baby boomers investing their assets across all styles and classes expecting to get about a 10% return over time with far less risk. Note that the back-tested returns do not include the 1970's where stocks and bonds both blew, mostly because that does not sell an investor on much of anything. This creates a perversion of risk far greater than what Jeremy Siegel started, where a large pool of investors can just sit on a diversified portfolio and with a little rebalancing earn 10% with minimal risk. Now I truly believe that diversification is a great thing, but going into a diversified approach based on biased past data is very dangerous. A diversified basket of US stocks and bonds with a hint of international and emerging markets is a great thing as long as they all go up. When they don't, and people's financial plan's don't go according to, well, PLAN, then this mark to model is then marked to market and a liquidity crunch arises. Does this sound familiar?
All of this risk perversion leads to the idea of risk inversion. Risk inversion is simply the idea that riskier assets are seen to be less risky than risk free assets. Any wealth manager uncomfortable with the market in January who allocated a client to cash for safety, is now rethinking that decision as the Fed has basically inflated equity markets and probably caused that safe cash allocation to be a losing bet. As this continues the moral hazard of a trigger happy Fed comes to light. Cash is no longer king and investors see risk free assets as quite risky, while those assets which are supposed to be risky as essentially risk free. I have been faced with this paradox lately and it is quite unsettling. I understand the dynamics of risk and return, but when the majority of risk involved is inflation risk, and inflation masks the actual investment risk, this is a phenomena that can continue for quite some time and indeed it has. What is more troubling is that we have a financial system based on the fact that investors are risk averse, but in being risk averse they become risk perverse and risk becomes inverted. Eventually, I believe we will reach a tipping point where the risk free asset receives a premium above the risk free rate of return - a paradox by many accounts. We started to see this this summer in the Treasury and Money markets as investors in short term treasuries and money market funds benefited from the panic and the "flight to quality".
Investors are constantly struggling with the tradeoff between growth, income, and safety of principal. Bond investors usually trade off growth for income and largely safety of principal. In turn they take on inflation risk. Equity investors usually trade off safety of principal for growth and sometimes income. In doing so they take on principal risk. In a healthy financial system, inflation risk should be less than principal risk (default and equity risk). We are currently in the opposite situation. The inflation rate is much higher than the default rate and principal risk is masked by inflation. This is risk perversion and risk inversion at its finest and the eventual inversion of this current valuation of risk will, in my humble opinion, be quite ugly.
ZB, your blog has moved to the "must-read" section of my bloglines account. I only have a hard copy (and it's at work, not here at home), but you should check out Edward Chancellor's "The Case For Cash" in the July (I think) issue of Institutional Investor.
A couple thoughts on "a better risk-free rate." It's too bad that TIPS auctions are not that big and usually only occur every couple months or so. Otherwise I'd advocate the use of a short TIPS bond yield as a better risk free rate, so we can disaggregate the real rate of interest from inflation. Then we can start getting over this lousy notion that stocks are better inflation hedges than real assets or even quasi-financial assets like commodity futures contracts.
Too many of the brokers-turned-planners' Asset Allocations only use mean-variance optimization and Markowitz Portfolio Theory. They assume a normal distribution of returns. The better planners use Monte Carlo simulations to calculate a Chance of Plan Failure. Why is this better? Because Jeremy Siegel can tell you the long-run average return on stocks, but can't tell you the *timing* of those returns. Try putting someone who retired March 10, 2000 in a room with someone who retired March 10, 1995, both with 100% stock allocations, and see if they're both happy with they're outcomes.
This is also why lifecycle funds are risky -- a recent Financial Planning magazine article showed that the average lifecycle fund with a 2010 retirement target date was just over 50% in stocks. Shouldn't that decision be dependent on the *size* of the nest egg as well as the time horizon? A multi-millionaire can afford to have more stock exposure than someone with $250,000.
Posted by: Bond investor | October 05, 2007 at 11:43 PM
Thanks for the thoughts and I appreciate your loyal reading and commentary. I enjoy writing this blog and at times it can be very difficult to balance that with work and life, so I am delighted when people like yourself are actually interested and inspired to comment and/or criticize what I post because my intent was to bring a different perspective to others.
I will try to get my hands on that article. I know a few people who subscribe to the publication. Stocks are only a good inflation hedge in the current environment that we are in. I think this has to do with a multitude of factors, some of which I have thought about during my response to this and may post on a little later. Essentially outsourcing and low interest rates along with growing consumer credit has created an inflationary storm that has made stocks a great investment for inflation hedging over this time period. I am skeptical as well as I know this is a phenomena that is not sustainable. Companies benefit from being able to keep input costs low through outsourcing and globalization. At the same time the cost of capital is low because of low interest rates, essentially. Meanwhile, large amounts of consumer debt has enabled companies to be able to raise prices and grow earnings. Consumer debt has replaced real income growth and has flowed through into earnings. If consumers did not have access to the debt, companies probably would either not be able to raise prices like they have unless they also paid there employees more - since many of their employees are also consumers of their products - in aggregate. Since extending credit increases the money supply it is no surprise that stocks have been a great inflation hedge over the past 10 or 20 years. What is scary however is that stock gains recently have mostly been inflation. Regardless of what exact principle you believe in, most exchange rate theory rests heavily on expected inflation. Todd Harrison over at Minyanville always reminds people that an investor in the S&P 500 would have lost money if he put money into the market in 2003, because currency losses outweighed the equity gains. I think this is beginning to unravel as debt is becoming more elastic and our inflationary measures are causing soaring prices all over the world.
As per the modeling, most big brokerage houses use Monte Carlo these days, but my argument is that it doesn't matter if you tell someone that we ran 10,000 trials and these are your odds, because the last 20 years of returns data is still introduced into the model, which happen to be the best 20 years of asset returns in the history of time. If you run a simulation model that simulates good times, you will get good times. Adding to that, these models take some sort of randomly generated numbers and then use them in some equation that gives investment performance based on the normal distribution, which Taleb and Mandelbrot (among others) have shown to be a poor proxy for past market returns. When I design a plan for a client or potential client, I make it as conservative as possible, putting in very modest return estimates, inflation expectations, and life expectancies. In my opinion, I believe this is going to be a huge deal for the industry in 10 or so years if things don't go well for the markets. I don't believe that these plans are however bad. I think they are a very good way to get people to think about the future, but are much more dangerous for both the clients and the industry if used improperly.
I hate the lifecycle funds as well. I tell everyone in them to get out of them. The allocations are arbitrary and usually are put out for a mutual fund to introduce a new product to put investors in their worst funds, and at an extreme, where everyone was in those funds, they would solely be driven by demographics. They also lull an investor into a false sense of security.
One asset allocation that I have been thinking about lately is what Taleb proposes in The Black Swan. Basically investing 80% of your wealth in safe investments like short to mid term treasuries and using the other 20% to make very speculative and very risk bets utilizing options or derivatives in some cases - arguing that people shouldn't try to get the average return - they should let their risk free capital be absolutely safe and their risky capital to be invested in very risky investments. As part of his Black Swan theory argues that it doesn't matter how good you are about picking it but by doing that you will put yourself into the position to profit from black swan events. It is interesting and alleviates risk perversion in general because people know what risks they are taking.
Posted by: Zero Beta | October 06, 2007 at 05:23 PM