Estimating a retirement income strategy requires a few assumptions including the expected return on invested assets. A common strategy is to use asset return data from the past, and specifically from the recent past in the United States (mainly because the data were readily available). Using this set of return data results in estimated safe withdrawal rates from retirement savings in the range of 4 to 5%.
Now, let’s imagine a world that looks nothing like the past: a world with more households in their peak saving years (between age 50 and 65) than at any time in the history of the United States; a world where America now makes up only 26% of the global bond market and less than 30% of the global market capitalization for stocks. This may not be a world that resembles the one that produced decades of high real yield for investors.
America has become a nation of yield addicts. When we’ve needed it, we’ve always been able to get a fix. Trust in a future with significant real yields and you can enjoy a secure retirement, a healthy pension, and you can even justify generous compensation for asset management services. But if it goes away, well, withdrawal is never pleasant.
It may be useful to revisit the purpose of financial markets. Financial markets exist to allow individuals and institutions to trade money across time periods and to transfer risk. Let’s for a moment ignore the risk part (although there is mounting evidence that the reward for taking it isn’t what it used to be). Trading money across time periods means that some people prefer to spend the money today and are willing to trade future dollars (through interest) with others who prefer to spend more in the future. This preference for consuming in the present rather than the future is the foundation of real asset yields. It means that when we give up spending today, we can spend even more in the future. That’s a good deal for those who place a lot of value on consuming in the future—for example retirees who need the money to maintain a lifestyle.
Retirees are now buying financial instruments (as are the institutions that control their assets) to fund a lifestyle for an expected longevity that will last longer than any post-retirement period in history. Imagine these retirees bidding for assets in a marketplace where participants over the age of 60 now hold 51% of the total investable wealth in the U.S. In addition to wealthy, older U.S. investors, assets are also being demanded by cultures with a taste for saving that is greater than our own. This scenario is beginning to resemble reality for many entering retirement today. And the implications on our addiction to real yield are sobering.
It turns out that there is an easy way to estimate the market’s willingness to trade consumption in the present for consumption in the future. There is an asset with almost no default or inflation risk that allows investors to purely invest a dollar today for a promised amount in 5, 10 or 30 years. That asset is the Treasury Inflation Protected Security, or TIPS. Recent auction prices for TIPS reveal a new reality that points to a yield-free world. For much of the 2000s, real yields on 10-year TIPS hovered around 1.5 to 2.5%. In September of 2010, they dropped below 1%. By November of 2011, they were zero and for each of the last five months have been negative. Investors were willing to give up an annual 21 basis point decrease in real consumption in order to transfer a dollar’s worth of spending today to a guaranteed, inflation-protected level of consumption in the future.
Negative TIPS rates may represent an overall market fear of inflation combined with a limited number of financial instruments available to serve as a hedge. However, May yields on 10-year Treasury bills are also running about 50 basis points below the April 2.3% inflation rate. It appears that investors have begun to accept zero or even negative real returns on safe assets over the coming decade.
Estimating withdrawal rate strategies without real yield is a gruesome task. So I asked my good friend, occasional co-author and withdrawal rate guru Wade Pfau, associate professor at the National Graduate Institute for Policy Studies, to calculate how zero real returns would impact traditional safe withdrawal rates. These estimates use the traditional methodology, which for better or worse assume constant inflation-adjusted spending from assets invested in either the S&P 500 or intermediate-term government bonds.
The retirement withdrawal rate literature used previous U.S. bond returns in order to estimate safe withdrawal rates. The real rate of return on intermediate term bonds between 1926 and 2010 was 2.52%. Using historical equity and bond return data, the failure rate of a 30-year retirement horizon is 6% using a 4% withdrawal rate strategy. However, as we know the best estimate of future bond returns is based on current market yields. If we really believe that future real bond returns will be 2.5%, then we should start a hedge fund that shorts long-run bonds. Financial advisors are better off assuming that the current market reflects reality.
When real rates of return on bonds are reduced to zero, Pfau estimates that the failure rate of a 4% withdrawal strategy increases the 30-year failure rate to 15%, or just over one out of every seven retirees. This near tripling of retirement default risk is disturbing, but it is not technically accurate because we also assume historical real equity returns. According to the capital asset pricing model, equity returns consist of an equity premium and a risk-free rate. Since the historical risk-free bond rate has been 2.5%, our historical equity returns are inflated if future risk-free returns are zero.
If we use a more accurate set of equity returns with a market-correct risk-free rate of return, the results are even more alarming. The failure rate of a 4% strategy with zero bond yield and a zero risk-free rate on equities is 34% over a 30-year time horizon. If real bond rates of return do not increase during a new retiree’s lifetime, they will have a greater than one in three chance of running out of money in 30 years. Even this estimation may be overoptimistic since current real bond yields are not zero—they are slightly negative. This estimation also includes an equity risk premium that existed during a period of equity returns so high that most economists consider them a puzzle. A more conservative (and more realistic) estimate of failure rates in a low-yield, low-risk premium marketplace would almost certainly lead to estimates of safe withdrawal rates well below 4%.
One easy way for retirees to gain yield on retirement investments is to receive a mortality credit that will give them a higher withdrawal rate on safe investments in exchange for a loss of liquidity and bequest by annuitizing their wealth. Many advisors bemoan current rates of return on annuities which are as low as 3.8% on an inflation-adjusted single premium immediate annuity. However, these rates are based on the current yield-free reality. Are you willing to gamble that the market is wrong?
Michael Finke is a professor and coordinator of the doctoral program in personal financial planning at Texas Tech University.