How you position clients’ high-yield exposures should vary depending on whether you think rates will go up or stay low. But what you should do in each case may not be as straightforward as you think.

High-yield guru Martin Fridson, CIO of Lehmann Livian Fridson Advisors LLC in New York, spoke with at the CFA Institute’s Annual Conference in Montreal last week to explain his take.

Assumption #1: Sluggish economy, rates stay low

Fridson thinks rates will stay low. “I’m more willing to take interest rate risk than credit risk right now. I’m not terribly concerned about rates going up because the [U.S.] economy continues to be lacklustre.”

Despite all the activity on the Fed’s printing press, we’re not seeing much inflationary pressure. One reason is globalization, says Fridson. “[It’s] preventing wage pressure—workers are not really in a position to demand large wage increases. And you’ve still got well under 80% capacity utilization in the U.S.”

Read: Anger over economy will make Trump president, says Gundlach

Weak economic conditions are paired with a negatively sloped yield curve for the high-yield market. “In other words, the risk premium and yields on 1- to 5-year paper are actually higher than on longer-dated paper,” explains Fridson.

He says we should expect this negatively sloped curve to persist for some time, and in such an environment, shorter-dated bonds will do better.

“For the retail investor, there are several short-duration, high-yield mutual funds, and if you own a conventional high-yield fund it’s worth considering switching to a shorter-dated one.”

Assumption #2: Rising rates

You’re in the minority if you think we’re on the cusp of a rising-rate environment. But if that’s your view, how should you position client portfolios? The conventional answer is to buy shorter-dated bonds, but available data doesn’t clearly support that view, says Fridson.

He notes there have been two periods where we’ve seen significant, sustained rises in government bond yields:

  • September 1998 to January 2000 (200 basis points)
  • May 2003 to June 2006 (261 basis points)

Shorter-dated bonds had much higher returns in the first period, which might make you think you’re supposed to buy these bonds when rates rise, notes Fridson. “But it turns out that was an unusual situation. At the beginning of that period, you had an essentially flat yield curve on government bonds, as opposed to the more usual situation where you have a positively sloped yield curve. So the shorter-dated bonds had an advantage.”

Read: Power up client portfolios with utilities plays

Turning to the second period, Fridson says you weren’t better off owning shorter-dated bonds. In fact, 15-year-plus bonds were the best performing. “That might sound very strange in a rising-rate environment, but that’s what happened. […] Here you had a positively sloped yield curve so the shorter-dated bonds didn’t have the advantage they had in that first period.”

Fridson argues that maturity data is ultimately inconclusive as a guide to figuring out how to position allocations for a rising-rate environment. “The best approach,” he says, “is by quality.”

There are essentially three divisions to the high-yield market:

  • BB
  • B
  • CCC and lower

In the 1998 to 2000 period, the lowest-rated bonds had the best performance, while the highest-rated bonds had the worst. The same pattern held in the second period.

You might say it’s a bad idea to buy lower-rated bonds in a rising-rate environment, since higher rates boost borrowing costs, potentially bankrupting those companies. But Fridson says that isn’t the case. “Companies have generally funded their maturities out, so the initial impact of an interest rate rise for many of these companies is zero. To the extent they refinance, it’s going to take years for that rise to entirely flow through to their borrowing costs.”

Read: Bond issuers behaving badly

He adds: “Typically, the reason rates rise is because the economy does better and there’s more demand for credit. And in those periods the riskiest companies tend to do best. The risk premium comes down because the economy is doing better, and in that environment, the more credit risk you took, the better [off] you were.”

So, if you think the U.S. economy is about to take off, “you would want to be overweight in the lowest-quality bonds—CCC and lower, though most money managers are not going to go below CCC,” says Fridson.

Selection by industry

Fridson’s analysis also provides guidance on which sectors to favour in a rising-rate environment.

In the 1998 to 2000 period, chemicals, metals and mining, steel, and telecommunications—cyclical industries that perform better when the economy is improving—did the best. Insurance also performed well.

“Not surprisingly, diversified financial services produced the worst return of any major industry during this period of rising interest rates,” notes Fridson. “Banks did better because the spreads on their loans tend to improve when rates go up.”

In the 2003 to 2006 period, chemicals, insurance, metals and mining, steel, and telecommunications “repeated as outperformers and could be considered for overweighting if you expect a rate rise.”

Healthcare and super retail (department, specialty and variety stores), which also did poorly in the first period, “were repeat underperformers and so could be considered as underweighting candidates in a period of rising rates,” concludes Fridson.

Read: China’s biggest problem—and how to fix it