How higher bond yields change the stock and bond mix in retirement

The yield on the 10-year US Treasury bond closed at 5.29 percent on September 30, its highest close since May 2002 and above its 2007 peak. A year ago it was 4.15 percent. For people in or close to retirement, the natural question is whether it now makes sense to keep more of their savings in bonds. This article runs that question through one retiree's plan, and the answer turns out to depend on why yields went up.

What happened to bond yields?

A bond's yield is the annual return you lock in if you buy it today and hold it until it matures. When yields rise, newly issued bonds pay more than older ones, so the price of bonds people already own falls until the two are roughly equal. Rising yields are good news for money you are about to put into bonds, and bad news for the value of bonds you already hold.

Over the past twelve months the 10-year Treasury yield went from 4.15 to 5.29 percent, and most of that increase came after July. The Federal Reserve also raised its own short-term rate in September, to a range of 3.75 to 4 percent, the first increase in three years.

One more detail matters for what follows. The Treasury also sells bonds whose payments rise with inflation, and the yield on those bonds is the return you receive on top of inflation. The difference between the two yields, known as the breakeven inflation rate, shows how much inflation investors expect, and that figure has barely changed: about 2.4 percent a year ago and about 2.4 percent today. Over the same period the yield on inflation-protected bonds rose from 1.78 to 2.93 percent. In other words, nearly all of this year's increase is in the return investors receive above inflation, and very little of it comes from expecting more inflation.

Why the reason matters for stocks

Over long periods stocks are expected to earn more than bonds, as compensation for the larger swings in their value. That extra expected return is known as the equity risk premium, and it cannot be measured directly; it has to be estimated, and professional forecasters often disagree about its size. When bond yields go up, there are two reasonable ways to think about what happens to stocks.

The first is that the expected return on stocks stays about where it was, so the gap between stocks and bonds gets smaller. Stock prices are some evidence for this view. When interest rates rise, stock prices normally fall until the expected return on stocks is again far enough above bonds, but the S&P 500 has risen about 2 percent since July and about 12 percent since the start of the year.

The second is that yields above inflation tend to rise when investors expect stronger economic growth, and stronger growth means higher company earnings. Under this view, stock returns rise by about as much as bond yields did, and the gap between them stays the same. That is an assumption, not a certainty, though it is a common one. Rising stock prices fit this view as well, since investors would be expecting higher earnings at the same time as higher interest rates.

It isn't possible to tell from market prices which view is right, so we ran both.

The retiree we tested

We used the same retiree as in our article on the 4 percent rule: a 65-year-old single woman in Florida, in average health, with $1 million saved. She has $600,000 in a 401(k), $300,000 in a brokerage account and $100,000 in a Roth IRA, and she keeps 60 percent of her savings in stocks and 40 percent in bonds. She spends $65,000 a year, rising with inflation, and claims Social Security at 65 for $24,960 a year.

We tested her plan under three sets of assumptions:

• Before the rise. The planner's standard assumptions, set in July when the 10-year yield was about 4.5 percent: bonds return 4.5 percent a year on average and stocks 8 percent.
• Bonds up, stocks unchanged. We assume bonds return 5.3 percent a year on average, matching today's 10-year yield in the same way the original figure matched July's, and stocks still return 8 percent. This is the first view above.
• Bonds and stocks both up. We assume bonds return 5.3 percent and stocks 8.8 percent, so stocks keep the same advantage over bonds that they had before. This is the second view.

Under each one we ran her plan at every mix from all bonds to all stocks, in steps of 10 percent, with 2,000 simulated lifetimes each time. The main result is the share of those lifetimes in which her savings last for the rest of her life.

What the results show

Here is her chance of her savings lasting, at every mix, under all three sets of assumptions:

At her current mix, her plan does better under either view: her chance rises from 68 percent to 71 percent if stocks are unchanged, and to 76 percent if stocks rise along with bonds.

The two views differ in what they say about the mix. Before the rise, she did better with each step toward more stocks, up to about 90 percent stocks. If bonds now return more and stocks do not, the curve flattens out: any mix from 50 to 90 percent stocks comes within about two percentage points of the best result, and a mix with only 40 percent in stocks gives her a slightly better chance (69 percent) than her 60 percent mix did before the rise (68 percent). If stocks rise along with bonds, the old pattern holds, with the whole curve sitting six to ten percentage points higher and her chances still rising as she adds stocks, up to about 80 percent.

The bad years

The lifetime figure only records whether the money lasts. It does not show how close the bad cases come, and for someone deciding how much risk to take, the bad cases are often the more useful thing to look at. Here is the chance that she still has savings at 85:

Before the rise, with 40 percent in stocks instead of 60, her lifetime chance was four percentage points lower and her chance at 85 was the same. Today, with 40 percent in stocks, her lifetime chance is two or three percentage points lower than at 60 percent, and her chance at 85 is higher under either view. The difference is clearest in the worst 10 percent of the simulations. With 60 percent stocks before the rise, she had already run out of savings by 85 in those lifetimes. With 40 percent stocks today, she still has about $13,000 at 85 if stocks are unchanged, and about $52,000 if they rise too.

For completeness we also ran a third explanation, in which yields rose because investors expect more inflation. That is not what the market data shows this year, and in that case she did not come out ahead at all: her chance at the 60 percent mix fell slightly, from 68 to 67 percent, because she would pay tax on larger gains without being any better off after inflation.

Should you hold more bonds now?

These numbers describe one retiree, and whether a shift toward bonds makes sense for you depends on what you need your savings to do. What the results do show is that she gives up less by choosing a more cautious mix today than she would have before this rise, under either view of the market, and that her worst outcomes improve more when she does. Whether that trade is worth it is a personal decision, and it helps to understand where the protection comes from.

A retiree who is living off savings is most exposed in the first years of retirement. If stocks fall sharply early on, she has to sell shares at low prices to cover her spending, and the shares she sells are no longer there when prices recover. Bonds fall less in a bad year, so holding more of them means selling fewer shares at a loss. That is why the bad cases above improve when she holds more bonds, and why the question matters most for people who are about to retire or have just retired. Someone who is still many years from drawing on savings has time for a fall in stocks to recover, and for that person the long-run growth of stocks usually matters more.

There are three costs to weigh against the protection. The first is that bond prices fall when yields rise. Anyone who already held longer-term bonds this year has seen their value drop, and if yields keep rising, bonds bought today will lose value too. The effect is larger for bonds that take longer to mature, so many people who want the higher yield without much of this risk stick to short and intermediate maturities. The second is inflation. An ordinary bond pays a fixed number of dollars, and at 3 percent inflation those dollars buy about 45 percent less after 20 years. Over long periods stocks have historically grown faster than inflation, though with large swings along the way, and for anyone worried about this, inflation-protected Treasury bonds now pay 2.93 percent above inflation. The third is tax. Selling stocks in a brokerage account to buy bonds can mean paying capital gains tax on the sale, while changing the mix inside a 401(k) or IRA does not, so one common approach is to make the change inside retirement accounts first.

What these numbers assume

In the planner, 5.3 percent is the average return on bonds for the rest of her life, which assumes yields stay around today's level; they may well go up or down from here. Her starting balance also does not reflect this year's fall in bond prices, since we held it at $1 million to keep the comparison clean. The planner assumes 3 percent inflation, a little above the 2.4 percent investors currently expect, so its figures for bond returns above inflation are on the cautious side. And this is one household, so your own numbers will differ.

The conclusion is narrower than a recommendation. Before this rise, holding 40 percent in stocks instead of 60 left this retiree with a lifetime chance four percentage points lower and no better chance of still having savings at 85. Today the same change leaves her lifetime chance two or three percentage points lower, and her worst outcomes are noticeably better. How much stock to hold is still a personal choice, but the cost of choosing the cautious side is lower than it was before the rise.