Maybe Past Performance Does Predict Your Savings' Future

Courtesy Pinnacle Advisory Group Close

Courtesy Pinnacle Advisory Group


Courtesy Pinnacle Advisory Group

You don’t have to examine America’s retirement safety net too closely to realize it has massive holes in it. Close to 60 percent of Americans have less than $25,000 in household savings, according to the Employee Benefits Research Institute, and 31 percent of retirees have less than $1,000. Meanwhile, financial advisers recommend having about 12 times your current salary saved to retire comfortably.

Even those who think they’ve saved enough may run out of money if they can't figure out how much they can safely withdraw from retirement savings each year. While a fixed 4 percent withdrawal rate was common in the past, financial advisers have been tweaking that formula. Michael Kitces, director of research at Pinnacle Advisory Group, has developed a system to determine withdrawals based on the stock market's price-earnings ratio. Lewis Braham caught up with Kitces at his office in Columbia, Maryland.

Q. Before we get into the nitty-gritty of withdrawal rates, let's go big picture. If you could implement any changes to improve the retirement security of Americans, what would you do?

A: There are a lot of behavioral nudges we can give people. Programs like automatic enrollment in 401(k)s have increased retirement account participation. There’s also the innovative "Save More Tomorrow" program created by Professor Shlomo Benartzi [UCLA Anderson School of Management]. This automates retirement savings for employees at an increasing percentage of their future raises. I’d like to see more of that.

Q: Why is saving a greater percentage of your salary as it increases important?

A: The rule of thumb has been to save a fixed percentage of income, say 10 percent or 20 percent. The problem is that when you systematically save a portion of your income, you also commit to increasing your spending with whatever's left over. So if I’m 28, making $48,000 a year, I’m saving $4,800 a year, or 10 percent. In three years my career ratchets up dramatically. Now I’m making $150,000. If I’m still saving 10 percent, the good news is I’m now saving $15,000 a year. The bad news is the $4,000 I’ve been saving in the past can’t possibly support a $150,000 lifestyle in retirement. So it’s better to commit to saving a much higher percentage of income as your salary increases.

Q: Got it. And now, how does using a withdrawal system that incorporates price-earnings ratios help retirees?

A: Traditionally advisers recommend a 4 percent to 4.5 percent withdrawal rate from your retirement assets regardless of market conditions. That kind of plan assumes the same rate of return regardless of whether stocks are overvalued, as they were in the year 2000, or undervalued, like they were in 1982. But my research indicates that if stocks are cheap, you can withdraw more from your account, as the future returns for your portfolio will be greater. And if stocks are expensive the opposite is true.

Q: What type of P/E ratio do you use to determine whether stocks are cheap?

A: The "P/E 10" is part of a broader group of valuation measures called cyclically adjusted price-earnings ratios, or CAPE. A lot of this research originates with Robert Shiller and his book, "Irrational Exuberance," where he found that a P/E ratio that incorporates 10 years' worth of inflation-adjusted earnings is remarkably predictive of future returns. By averaging the earnings of stocks over that period, you smooth out market cycles so profits are neither inflated nor depressed. CAPE works remarkably well in predicting 10- to 15-year returns. And the longer-term return is what drives withdrawal rates.

Q: According to your research, if the P/E 10 is above 20, the withdrawal rate should be 4.5 percent of assets. If it's between 12 and 20, the rate goes to 5 percent, and if it goes below below 12, then you can withdraw 5.5 percent of assets. How did you determine those ranges?

A: That's literally what I found in the historical data when I ran simulated portfolios dating back to 1871. [The portfolios were 60 percent stocks, 40 percent bonds and assumed a 30-year retirement holding period.] These withdrawal percentages historically have ensured that the portfolio did not run out of money. So long as the P/E 10 ratio was below 20, the standard 4 percent to 4.5 percent withdrawal rate proved unnecessary. Because no matter what catastrophic historical event happened to the markets, you eventually would get enough return if you started retirement when valuations were in that range.

Q: What is the P/E 10 ratio now?

A: It’s in the 24 to 25 range. That is about as high we’ve seen historically. Heading into the Great Depression, it was 29. Going into the 2000 tech crash, it was in the 40s. We’ve certainly gotten higher, but we are in the bad zone. So withdrawal rates should be on the low end right now.

Q: How do the current low bond yields affect this?

A: The short answer: It certainly is not good. In many historical periods, favorable returns on bonds helped offset bad returns on stocks. Certainly it’s concerning, but at some point these things start to interact with each other as well. Sustaining ultra-low bond yields can push stock prices up. But an increase in inflation and interest rates will cause a reversal of that trend as bond prices fall. A bond bear market can foreshadow an even worse stock market.

(Lewis Braham is a freelance writer based in Pittsburgh.)

To contact the editor responsible for this story: Suzanne Woolley at

Press spacebar to pause and continue. Press esc to stop.

Bloomberg reserves the right to remove comments but is under no obligation to do so, or to explain individual moderation decisions.

Please enable JavaScript to view the comments powered by Disqus.