From the October 2016 issue of Research Magazine • Subscribe!

Finke: Keys to Retirement Planning in a Low-Yield World

It’d be nice to believe that market returns in the 21st century will be as high as they were in the 20th. The science suggests that they won’t.

The 1990s can be seen as the golden era of investment management. The rising tide of stock prices lifted all boats — even the silliest skiffs. Goals were easily met. Clients were happy. Advisor fees seemed modest.

But what if I told you that the party of the 1990s was a fool's paradise? What if the outsize returns of the ‘90s were a necessary precursor to a new era of more modest asset returns in the 21st Century? And what if the salad days don't return? What if the era of low returns is here to stay?

Researchers in financial economics have been gathering data that suggests we are in fact entering a new low-return environment for both safe and risky investments. The data is convincing and, if true, the impact on the investment advising profession will be enormous.

What are asset returns? It's surprising that professionals trained in the economic sciences don't stop to think about financial assets as normal economic goods subject to the forces of supply and demand. In reality, there is a finite supply of financial instruments. And when demand goes up, so does price.

What happens to returns on these assets when prices rise? They go up as well — until they don't. Then they go down. Consider this fact pointed out by Harvard finance professors Robin Greenwood and David Scharfstein.

Between 1980 and 2007, the market value of stocks grew from 50% of U.S. gross domestic product to 141% of GDP. This would be great news, if companies in 2007 were three times more profitable than they were in 1980. Unfortunately, it wasn't an increase in profits that drove the increase in stock values. It was an increase in prices. Investors were willing to pay 266% more per dollar of earnings in 2007 than they were willing to pay in 1980.

The implications are sobering. Stocks rose in the 1990s, because investors were willing to pay more per dollar of earnings (which can be either paid back in dividends or reinvested for growth). Investors received fantastic returns in the late 1990s, since the demand for stocks shifted outward. And when demand for stocks goes up, expected returns on stocks (after prices have risen) goes down.

Another interesting implication of the rise in stock values between 1980 and 2007 is that nearly all of the revenue growth in investment management that occurred during this time can be attributed to higher stock prices.

This is worth repeating. For those who charge a percentage of assets under management, your revenues have tripled between 1980 and 2007 because the price of stocks went up. But if your clients only received returns on the dividends paid from profits in these public companies, their returns would not have tripled. They wouldn't have gone up at all. They are paying three times as much for each dollar of stock earnings. And you, the advisor, are receiving three times as much revenue from the client for each dollar of corporate earnings.

Of course, stocks aren't the only financial asset that has become more expensive as demand for financial assets rose during the past three decades. Bonds are also significantly more expensive. Consider that as late as Jan 1, 1990, clients would need to invest $122 to receive $10 in bond income from 10-year Treasuries.

Today, an investor needs $641 to buy $10 of bond income. That means a client working with an advisor charging 1% would pay 12.2 cents for each dollar of bond income. Today's client pays an advisor 64.1 cents for each dollar of bond income.

In 1980, a client paid $88.50 to receive $10 in stock earnings. Today, a client must pay $270 to get $10 in profits. That amounts to 8.85 cents for each dollar of earnings in 1980 and 27 cents for each dollar of earnings in late 2016. And that doesn't include any fund or transaction expenses on the assets themselves.

If two-thirds of a client's growth in safe assets and one-quarter of growth in risky assets are eaten up in asset management fees, then advisors face significantly greater pressure to justify their value beyond producing investment returns. This may be less important if returns rise again, but the research on future returns isn't reassuring.

More Bad News for Future Returns

Why are investors willing to pay more for each dollar of stock earnings and bond income? What is the evidence that demand for financial assets is rising?

First, let's consider the source of stock prices. The traditional model explains the price today depends on the risk-free rate plus the risk premium. What is the risk-free rate today? A 3-month T-bill's yield hovered between 5 and 10 basis points (0.05% to 0.10%) between 2011 and 2015. The historical average between 1934 and 2015 was 3.6%. So the expected return on stocks is already starting 3.5% behind the historical average.

Now, let's move to the so-called equity risk premium. The equity premium is the amount of extra return needed to compensate investors for holding an asset with a payout that could be uncertain. Most of us look to history to determine how much extra return investors hold for accepting risk.

The size of the risk premium depends on which period of history you examine. In a 2002 article titled “The Equity Premium,” by Nobel prize-winning economist Eugene Fama and Ken French, they estimated the expected risk premium from dividend and earnings growth and found that stock prices rose more than would be expected during the latter half of the 20th century.

Why? “The high return for 1951 to 2000 seems to be the result of low expected future returns,” they said. Stock prices rose more quickly than dividends or earnings, because investors didn't require such a high rate of return from stock investments.

Why don't investors require the same premium in order to hold risky assets? Fama and French attribute this to a combination of factors including much wider stock market participation by both individuals and institutions (one in seven Americans owned stock in 1980 compared to half of Americans by 2000) and the emergence of mutual funds that allowed investors to hold well-diversified portfolios at a low cost.

We can also add a big reduction in the cost of buying stocks after the May Day deregulation of the brokerage industry in 1975 and a decline in capital-gains tax rates. In an article published in 2008, co-authors from New York University and Wharton attribute the recent rise in stock prices to an increase in global macroeconomic stability. Who can doubt that an investor in 2016 should feel more comfortable buying risky assets than an investor in 1930 or 1940?

The consensus is that stock prices are higher now since investors don't need as much of a premium to get them to switch from bonds to stocks. The current Shiller 10-year trailing price/earnings ratio for the S&P 500, which stands at 27, is about 75% above the historical price investors are willing to pay for a dollar in corporate earnings.

What does it mean that stocks are more expensive as a multiple of real earnings? Consider an investor buying stocks when the market has a P/E of 16. For each $100, the firm produces $6.25 in earnings. Of these earnings, they can either return the cash to investors through dividends or they can reinvest in projects that will provide future growth.

At today's P/E of 27, each $100 buys $3.70 in earnings. About $2 of this is currently being paid back to investors in dividends (or a 2% return on investment). Of the rest, $1.70 is being plowed back into the firm for future growth. That is about half the amount reinvested per $100 of stock price when P/E ratios are closer to their historical average.

In order for stocks to have any hope of approaching their historical premium (don't forget that we’re already starting with a risk-free rate 3.5% below the historical average), the growth rate on these reinvested earnings would need to be double the historical growth rate on reinvested capital.

That's not likely to happen and is one of the reasons why there appears to be a nearly linear relation between the price of stocks and the subsequent 10-year returns. In a fascinating 2002 article, AQR Capital co-founder Cliff Asness sorted stocks by their price using P/E ratio and tracked them over the following 10 years. He found that subsequent return fell consistently at each increasing price level. When stocks were as expensive as they are right now, the average return was 0.5% above the rate of inflation. The highest historical return was 6.3%, and the lowest -6.1%.

Since today's inflation-protected bonds pay about 0% return across industrialized countries, and high stock prices suggest a modest equity premium of perhaps 2 or 3% at best, reaching long-term goals will be much more expensive for today's investor. And for those who still believe that returns will eventually go back up, you should take no comfort in the significant drop in both bond and stock prices that will be needed to boost future bond and dividend yields.

Low Returns and Planning

Advisors accustomed to a world of 10% returns on stocks and 4% returns on bonds may be surprised by how sensitive common planning strategies are to a low-return environment. In 2013, my co-authors David Blanchett and Wade Pfau and I published a series of articles detailing how more modest returns on stocks and bonds impacted the so-called 4% rule in retirement.

When we ran Monte Carlo simulations that randomize stock and bond returns at their historical average, we found that the 4% rule has about a 95% chance of working. That is, a retiree can safely withdraw an initial 4% of his retirement savings balance and increase spending at the rate of inflation over a 30-year time horizon.

When we use today's low bond yields and project a random return upward toward the historical average (this may be optimistic) and use projected stock returns that incorporate current valuations, we find that the failure rate jumps from 5% to close to 50% for a balanced portfolio.

One of the reasons failure rates are so sensitive to asset returns is that low rates mean a retiree will have to spend down investment principal in order to meet a spending goal. Historical asset returns meant that the bond portfolio earning 4% could nearly sustain spending in the early years of retirement and high equity returns were just icing on the cake. Today, the bond portfolio is depleted each year, and the stock portfolio is expected to be just as volatile in the future but have a lower average return. It is less likely to bail out a bad sequence of returns.

So a traditional investments-only strategy won't sustain a 4% spending goal in a low-return environment. What are the other options? First, you could ramp down the spending goal to 3% to reduce the failure rate closer to 10%. But that isn't the only option.

One of the most effective ways to get a higher return from safe assets is through mortality pooling — otherwise known as buying an income annuity. A deferred income annuity will provide the biggest bang for the buck.

For example, an advisor putting together a bond ladder to fund later life spending can buy the same income from the ages of 80 through 100 at half the cost through a deferred income annuity. That's because the so-called mortality credits that a retiree gains from pooling longevity risk with others rises with age.

Another option is to maintain some spending flexibility. One easy method is to follow a so-called required minimum distribution, or RMD, spending rule that mimics the rules used by the IRS for required withdrawals from IRAs. The RMD rule is helpful, because it bases spending on current assets and expected longevity.

One of the problems with the 4% rule is that it sets a spending goal the first year of retirement and ignores subsequent portfolio returns and longevity. The RMD rule incorporates both the current portfolio size and longevity. A good rule of thumb is to add two to four years to the Social Security average to ensure spending won't have to fall too much later in retirement.

A low-return environment affects lifestyle expectations across all ages. If younger clients can't expect to earn as much on investments, this means they will both have to save more and change attitudes about their goal retirement age and lifestyle.

Using more realistic projections of stock and bond returns, a client may have to save so much in order to reach a goal of retiring at 62 that they’re better off saving a little less and retiring at 67. For advisors who are using software that allows them to modify investment projections, adjusting down expected returns will give clients a clearer understanding of these tradeoffs.

It would be nice to believe that stock and bond returns in the 21st century will be as high as they were in the 20th century. The science suggests that they won't.

Advisors need to help clients adjust portfolio return expectations and be honest about their own limited ability to meet ambitious return targets. And none of the implications is helpful for the industry. Lower account balances, greater pressure to justify fees, and the possibility of disappointing clients who need to face a more modest financial reality are the price we pay for living through the golden age of investment management.

--- Related on ThinkAdvisor:

Page 1 of 4
Single page view Reprints Discuss this story
We welcome your thoughts. Please allow time for your contribution to be approved and posted. Thank you.


Finke: For Retirees, Spending Too Little May Tarnish Golden Years

Though many people fear they will run out of money in retirement, the reality is that most don't spend enough...

Most Recent Videos

Video Library ››