The New York Times has noted that the US is heading toward an inverted yield curve, which is one of the precursors for a recession. The yield curve is said to be inverted when long term government bonds pay less interest than short term ones. I'm confused at why the article would predict that people would invest in the long term bonds right now:

If enough investors begin to grow concerned about a recession, they will most likely put more and more money into the safety of long-term government bonds. That buying binge would likely help flatten, or invert, the yield curve.

Maybe my reasoning is oversimplified, but isn't it smarter to invest in the bonds that pay more interest so you make more money? In this case, you should invest in short term bonds because they will pay more once the yield curve inverts.

5 Answers 5


Short-term bonds might pay more interest, but only for a short time. When they mature, overall interest rates available for reinvestment are likely to be lower in this scenario. (Long-term rates are based largely on expectations of future short-term rates.) By investing in long-term bonds, you lock in a rate that you see as still relatively high, and express your opinion that we are headed for a lower-return environment. This view prevailing is what causes the inversion.

EDIT: Moreover, even if you don't plan to hold the long-term bonds to maturity, you can benefit from their "locked-in" rate in another way: If overall rates fall soon, as you expect, then the price of your long-term bonds will rise and you can sell with a capital gain.

  • So these people are locking in the long term rates BEFORE the yield curve inverts, assuming that long term rates will continue declining. But since we're already pretty close to an inverted yield curve, isn't it better to wait when the curve goes back to normal and the long term rates skyrocket? In simpler terms, you want to lock in a high rate. Why lock it in when the rate is already very close to short term rates?
    – JoJo
    Jul 1, 2018 at 21:01
  • 5
    @JoJo Whether today's rates are "high" or "low" will only be known when we see what the future brings. If long rates were likely to "skyrocket" sometime soon, then indeed it wouldn't make sense to buy long-term bonds now. But it is also possible that both long and short rates could decline for an extended period. Even after an inversion happens, you would buy long bonds if you think rates will fall more than the market expects. A normal curve implies a consensus for stable or rising rates; an inverted curve implies a consensus for falling rates.
    – nanoman
    Jul 1, 2018 at 21:58

Bonds are interesting and you've come across most of what makes them so interesting. The interesting thing with bonds, is all the already outstanding notes of similar remaining duration and underwriting risk are revalued to the newest yields. The "why" will shake out like every single other investment decision. Does this make sense for my portfolio?

I'm sure you know this part but I'll start at the beginning. A bond is a note that says, for $1,000 of your money today I will pay you $1,010 a year from today. The bond issue might be for $1,000,000; so there are 1,000 of these notes. Lets say that they raise the million they are looking for, 1,000 people show up with $1,000 each. Everyone gets a 1% yield; $10 return on $1,000. But lets say that at auction two only $990,000 shows up. These auction two notes are bought for $990 each, they'll all still pay $1,010 after a year, these have a yield of 2%.

If all the holders of auction one hold their notes to maturity, they still realize the 1% yield that was acceptable for their money at the time they bought. But if someone wants to get out of their auction 1 note, they'll have to sell it at the now current yield of 2%, thus taking a loss. Who would buy your auction one note for $1,000 when they could get someone's auction two note for $990. This effect cascades through the varying time-frames and yields. A 30 year note with one year remaining will be priced to a yield of the other one year notes, not the 30 year notes.

Whoever buys the note at the time of issue has decided, x% return on my money is acceptable for that given time frame. Increasing or decreasing interest rate environment doesn't matter. At the time of issue at auction one, 1% was enough given the facts and circumstances.

SHOULD you buy whatever bond if there's a strong possibility we're in an increasing interest rate environment? That depends on your investment horizon. Ideally, when you buy a bond, you are committing that money FOR THAT TIME PERIOD. If rates raise and the value of your bond drops, it doesn't matter to you because your bond will still pay you your agreed upon yield at the time of maturity.

So, who is buying 30 year government bonds? Someone with a lot of money that wants to guarantee a x% return over that time period. Should YOU buy 30 year treasuries? That depends... do you have a 30 year horizon over which you'd like to lock x%?

As to the indicators of recession, you should structure your assets based on what your best understanding of the market. But you should never buy an asset with a longer time frame than your investment horizon.


Say that you want to lend money for a ten-year period. The yield curve bakes in expected future short rates, so it's always shaped so that you're indifferent between

  1. rolling over one-year bonds for ten years
  2. buying a ten-year bond

If the future path of one year rates ends up being higher (lower) than the market expects, then the first (second) strategy is better.

For an extreme example, say one-year bonds yield 2% and ten-year bonds yield 1%. Suppose you follow the first strategy, but the next day, a disaster happens, and one-year rates fall to 0.0% for the next ten years. Then you'll wish that you'd locked in the 1% yield for ten years.


but isn't it smarter to invest in the bonds that pay more interest so you make more money? In this case, you should invest in short term bonds because they will pay more once the yield curve inverts.

If the yield curve is inverted, it means people think that in the future yields will drop.

So you can get a high(er) interest rate now for a short term bond, but when it expires you'll (the market believes) only be able to re-invest at a lower rate.

If you take the lower rate on the longer term bond now, you'll lock in that rate even if rates drop.

As a crude example, if short term rates are currently 2%, but the market thinks they will drop to 1% next year, then 5-year bonds might pay 1.5%.

The reason that this relates to predictions of recession is that central banks often reduce interest rates in response to a recession. So if you think a recession is likely, then you would expect interest rates to drop, and be willing to take a long term bond at a lower rate than a short term one.


Suppose interest rates are 5% on short term bonds (1 year) and 4% on long term bonds (10 year).

Now, suppose you predict that (short term) interest rates will fall to 2% in a year, stay there for 5 years, then recover 0.5% per year up to 5%. This is consistent with a hard recession with a slow recovery.

The long term investor gets 48% return over 10 years.

The repeated short term investor gets 5%, 2%, 2%, 2%, 2%, 2.5%, 3.0%, 3.5%, 4.0%, 4.5%, 5%, for a total of 41.7% return over 10 years.

Buying that long term bond, even if it gave less return, let you keep that return year-over-year while interest rates where low.

There is a downside, in that your money is locked up in a bond; but bonds themselves can be sold, and one with a higher rate than what you can currently get from the government sells at a premium.

When short term rates actually fall, long term rates tend to follow; but your locked-in rate won't. So if you are right about the recession, you can sell your 9 year 4% bond next year for more than 1.04x the money you put into it initially (in effect, move the return up).

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .