After 70 Years of Fruitful Research, Why Is There Still a Retirement Crisis?
In celebration of the 70th anniversary of the Financial Analysts Journal, CFA Institute published a special issue focusing on retirement security. The following is an excerpt from the guest editorial written by Laurence B. Siegel, the Gary P. Brinson Director of Research for the CFA Institute Research Foundation.
Despite thousands of scholarly and practical articles (many of which have appeared in the Financial Analysts Journal over the years) and much earnest effort by researchers, financial product designers, pension plan sponsors, advisers, legislators, regulators, and individual investors, we still have a retirement crisis.
The contents of this special retrospective issue on retirement strongly suggest that we have both the intellectual tools to avoid a retirement crisis and many (not all) of the needed institutional arrangements.
Richard Bookstaber and Jeremy Gold’s 1988 article “In Search of the Liability Asset” shows that the pension liability contains not only bonds (the obvious choice) but also equities. The presence of equity in the liability comes from the sensitivity of the liability to economic growth and wage growth in the long term. Although their article was originally intended to apply to defined benefit (DB) plans, in the new defined contribution (DC) world the authors’ argument becomes central, because the saver’s portfolio is set against her whole economic liability and thus should usually contain equities as well as “safe” assets.
Time diversification is the idea that risky assets, such as stocks, are less risky if held for a long time than if held for a short time. If time diversification exists, then retirement investors, who don’t need their money back for a long time, can engage in a kind of time horizon arbitrage against short-term investors, whose aversion to price fluctuations causes the equity risk premium to be high
In his 1994 article, Mark Kritzman shows that under simplified conditions, this arbitrage does not exist. Long-term investors (including retirement investors) face at least as much risk from risky assets as do short-term investors.
Writing near the peak of the bull market in the late 1990s, Peter Bernstein studied past returns and valuation levels to determine what “basic” or repeatable rate of return would have been earned in the stock and bond markets if valuation levels had not changed. He showed that stocks earned an arithmetic mean real annual “basic” return of 5.7% whereas bonds earned a “basic” return of 2.7%.
Bernstein thus concluded that the equity risk premium (over bonds, not cash) is 3%, lower than the estimate reached by many other authors but still positive to an economically significant degree. Then as now, knowing what returns to expect from each major asset class is critical to deciding how much to save, what assets to invest in, and how much to spend.
In Zvi Bodie’s 2003 article, his main point, which has influenced almost all subsequent retirement research, is that “lifetime consumption of goods and leisure,” not “end-of-period wealth,” is what retirement savers care about. Consequently, for those saving for retirement, the riskless asset is not cash — as Harry Markowitz would have it — but, rather, a portfolio of laddered bonds or TIPS (Treasury Inflation-Protected Securities), with each cash flow from the portfolio matched to a consumption need.
In his article, Bodie also suggests two financial products that, if properly developed, would make the provision of retirement income easier: an escalating life annuity and a product that bundles together longevity annuities, which appeal to the healthy, and nursing home insurance, which appeals to the sick. By offering both types of protection in one product, the adverse selection that affects both products is mitigated and many more people are covered (and at lower cost).
Keith Ambachtsheer suggests in his 2007 article that the “DB or DC” dilemma is misstated and that “neither DB nor DC” is the path to a better pension system. The system he favors, modestly called TOPS (The Optimal Pension System), involves DC-like contributions that are used preferentially to buy life annuities but that can also be used to make non-annuitizing investments. If this sounds a lot like TIAA-CREF, the US private pension plan for college teachers, it’s intentional: Ambachtsheer proposes that organization as a model, with some modifications, for a universal retirement system.
Don Ezra, also writing in 2007, begins by telling the true story of how the DB system was destroyed. As one might guess, the near demise of the DB system was not intentional; it was the consequence of a set of historical accidents, coincidences, and misbehaviors. Ezra then presents an intriguing DB plan design for the future. Fully funded by law, Ezra’s plan is “nonpenalizing” in the sense that short-service employees get as large a pension benefit per employer-contributed dollar as do long-service employees. Thus, the system is portable — and fairer than in current practice.
Ezra also suggests a fix for DC plans: They should use auto-enrollment, auto-escalation, and auto-conversion into annuities, either at retirement or upon achieving one’s life expectancy. The latter choice makes it possible to capture much of the benefit of annuitization without losing all of one’s liquidity at retirement.
Annuities! Everybody loves them on paper — they replace DB income streams and capture the huge gains from pooling longevity risk. But most investors shun annuities today, given their inflexibility, high fees, adverse selection, and lack of a transparent market.
Observing that almost no one puts all his wealth into annuities — nor should he — Jason Scott’s 2008 article developed an optimization method for deciding what annuities to buy and how much. The answer is shockingly simple. The key is to break up an annuity promise into year-by-year promises — income in one’s 65th year, 66th year, and so forth.
Although much advice to investors is rendered without considering taxes, “taxes exist,” as William Reichenstein, Stephen Horan, and William Jennings remind us, and can consume a huge slice of retirement savings. The authors’ analysis in this 2012 article produces an asset allocation that does not ignore taxes. Moreover, the analysis produces a substantially different asset location, the practice of holding each asset in the type of account that produces the highest after-tax present value. Investors, advisers, and plan sponsors who are not paying attention to asset location had better start.
How much do you need to save for retirement? Many would-be retirees seem baffled by this question, because it’s hard — or impossible — to forecast market returns. By assuming riskless investing, Stephen Sexauer, Michael Peskin, and Daniel Cassidy in their 2012 article have eliminated the need to make such a forecast and have arrived at a “retirement multiple” that enables investors to set an asset accumulation target. To keep the multiple from growing unmanageably large, the authors assume that retirement income after age 85 will be funded by a deferred income annuity.
In a new article written for this special issue, Barton Waring and Laurence B. Siegel show that each year one should spend (at most) the amount that a freshly purchased annuity — at then-current portfolio values, interest rates, and number of years of required cash flow remaining — would pay out in that year. Investors who behave in this way will experience consumption that fluctuates with asset values, but they can never run out of money.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.