Felix Salmon

Do TIPS ETFs make sense?

By Felix Salmon
August 18, 2009

It’s quite clear that Pimco is talking its book when it says that active management makes more sense than indexing if you’re buying TIPS. On the other hand, Pimco’s arguments are quite compelling — more compelling, indeed, than counterarguments which don’t really address what Pimco says and instead fall back to the standard passive-investing-is-always-better approach.

The WSJ came to exactly the right conclusion, I think, after looking at the numbers:

Over the past five years iShares Barclays TIPS Bond posted an average annual return of 4.28%, according to Morningstar. That is slightly better than the 4.14% return on the Pimco Real Return fund’s “D” shares, which investors can purchase through a discount brokerage account.

But as always with mutual funds, costs are key. Investors with access to the Pimco funds’ “institutional” shares through a large company retirement plan would have outperformed the ETF, while those in one of the load-paying broker-sold shares classes would have done significantly worse.

I’d only add that it’s crucial, whenever you’re dealing with TIPS, to understand the tax consequences of investing in them — that can make a huge amount of difference, and it’s not uncommon for investors to avoid TIPS entirely just because they can’t deal with the concomitant tax hassle.

As for bond investing, there’s an extra layer of complication here in that it’s very hard to come up with a good index: you can’t just buy-and-hold bonds, as you can with stocks, because that way your duration is constantly declining. As a result, it’s far from clear what constitutes outperforming or underperforming — everything just becomes relative to other investors’ performance.

Still, I’ll stick to the ETFs, just because they’re easier to understand and it’s less likely I’ll be ripped off or subject to some managerial blow-up. My gut feeling is that individual investors are generally foolish to seek outperformance — which is ultimately what Pimco is selling here. And that anybody who can outperform a little can underperform a lot.

6 comments so far | RSS Comments RSS

Why not just buy individual TIPS?

Posted by fusion | Report as abusive


(1) as mentioned, you have to self-manage the constantly changing outstanding duration of one or many separate bonds. However, laddering would help, be close enough, and not too difficult.

(2) TIPS bonds owned directly, outside of a tax-deferred account, have tax implications that can be less than optimum (taxes on unrealized gains due to inflation adjustments). (but see http://www.frbatlanta.org/filelegacydocs  /wp0309a.pdf) ETFs can shield/defer that.


Don’t forget Vanguard VIPSX, expense ratio .24% (or its VAIPX Admiral Shares .15%). I believe you can get both Pimco and T.Rowe Price TIPS funds without a load, but the expenses are .74% and .50 respectively, which seems a bit high for managers who consistently underperform Vanguard’s offerings.

As far as taxes, the problem isn’t as acute as you make it out to be. When investing in bonds directly, it is true that you are taxed on the inflation adjustment to the principal, even though the adjustment only goes into principal. It’s not tax efficient in that sense, because you are paying taxes on cash you don’t receive until you have sold the bond. It’s annoying, but I wouldn’t call it a “hassle” unless one is totally strapped for cash. However, bond funds do distribute both the inflation adjustment and the coupon quarterly, which solves the cash flow problem. In any event, these problems disappear if you have your TIPS in an IRA, which is probably where they should be anyway.

Posted by maynardGkeynes | Report as abusive

(1) I’m happy with a constant stream of real income, with a guaranteed real principal payment at the end. Why should I care about changing duration if I won’t sell?

(2) TIPS in taxable accounts are no worse than nominal treasuries. Better if there is unexpected inflation. The trouble is you could lose on an after-inflation, after-tax basis if you hold in a taxable account. Holding in a tax-advantaged account is much better.

The more interesting tax comparison is to short munis, which should adjust for inflation and which don’t lose to taxes. If they adjust enough, they beat TIPS after tax and inflation if there is high unexpected inflation.

Posted by fusion | Report as abusive


“I’m happy with a constant stream of real income, with a guaranteed real principal payment at the end. Why should I care about changing duration if I won’t sell?”

Good question.

I’m sure Felix or commentators can jump in with maths, but the reason why average bond duration matters goes something like this: if your TIPS bond expiration is exactly matched up to your desired inflation-protection period (and no ante- or post- considerations), then you are correct: hold-to-maturity = the inflation protection you imagined.

But if those periods differ: imagine that the value of the embedded inflation protection option varies over time, and you are forced to do a renewal market transaction at a high or low point on that curve, not when you want to relative to your entire portfolio: volatility in cash flows means that actual inflation-protection at any given time does not equal the average inflation protection desired over the time period.

I would guess that terms like gamma-hedge and delta are involved, but can’t speak to all of that business. But I’d be psyched if it turns out that my high-school calculus has something to do with it, because I just thought all that up.

“TIPS in taxable accounts are no worse than nominal treasuries. Better if there is unexpected inflation. The trouble is you could lose on an after-inflation, after-tax basis”

Even better than that, so I just learned: “Are Treasury Inflation Protected Securities
Really Tax Disadvantaged?” http://www.frbatlanta.org/filelegacydocs  /wp0309a.pdf

“We find empirical evidence that the after-tax yields on TIPS and conventional bonds are close to one another when inflation expectations are taken from the two markets.”


1) Owning a fund would not seem to solve that problem

2) I mentioned inflation expectations above. The result really should not be surprising (assuming you have the cash available to pay tax on the phantom income). The problem with the article is that it ignores other alternatives.

Posted by fusion | Report as abusive

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/