And why do economists think it’s a precursor to a recession?
If you lend me money, I pay you interest.
Long-term loans are riskier than short-term loans, so you will insist on a higher interest rate.
When people start offering or accepting higher rates for short-term loans than for long-term loans, it indicates trouble.
Inverted yield curves do not always lead to recession, but recessions are nearly always preceded by inverted yield curves.
Bonds have “maturities”.
When the bond matures, the money is to be paid back. Some bonds come due in one month’s time, other bonds are paid back after a decade.
Normally, the bond with a short maturity has a lower yield-to-maturity – a lower interest rate – than the bonds that have long maturity. Which makes a certain amount of sense, yes? If I lend you a thousand bucks, to be paid back next month, then maybe you have financial problems and can’t pay me back (or for government bonds, maybe the inflation rate spikes in the next month and you pay me back with dollars that aren’t as valuable) but that isn’t necessarily likely. But if you pay me back the principal of the loan ten years later, well shit, a lot of things can happen in those ten years. So the loan is likely to made on stiffer terms, with a higher yield-to-maturity.
A yield curve is a list of all the bonds’ interest rates, stacked up by how long each matures. Normally, this is an increasing curve. Look at the wik picture for an idea.
An inverted yield curve is when the interest rate on some of those long bonds, to be paid back many years from now, is lower than the interest rate on the shorter bonds. The bonds that are to be paid back soon have higher yields than the bonds that won’t be paid back for a while.
This is weird.
Statistically, an inverted yield curve often (though not always!) comes before a recession. That gets much more complicated.
It seems to me that this wouldn’t actually be a predictive indicator. If the yield curve is inverted, it just means that a lot of people think that there’s a recession coming, based on whatever other indicators they’re following. In other words, anyone offering such loans, you could ask them why they’re charging more for short-term loans than for long-term ones, and they could give you some sort of answer.
There isn’t really anyone “offering loans” to ask. Bonds have a fixed set of coupons and a fixed payment at maturity and the yield is set by offering the bond at auction and getting the implied yield from there.
E.g. you could be offering a note that says “Whoever holds this note I will pay $50 every year for 10 years and $1000 at the end of the 10 years” and sell that note at auction. If that note sells for exactly $1000, the rate is implicitly 5%. If it sells for less than that it is more than 5%, and if it sells for more than that, the rate is less than 5%.
The point is, bond yields are determined by bond prices, and bond prices are determined by a market just like stock prices are, and there is not necessarily anyone buying or selling both 2 and 10 year bonds to ask why they’re doing that.
But it’s still the result of decisions people are making. If yield curves are changing over time, then someone must be making different decisions. Someone’s buying bonds at different prices, or someone’s buying them now who wasn’t before, or not buying them now who was, or probably a whole big mixture of all of those. “The Market” is just a collectivization of a whole lot of individuals making decisions; it’s not a decision-maker in itself.
Certainly not perfectly predictive.
There have been plenty of times that an inverted yield curve was not followed (quickly) by a recession.
We’re talking about sovereign lending here, not individual loans. The US Treasury market is one of the most liquid markets in the world.
One of the underlying rationales of a large liquid market is a “Wisdom of Crowds” sort of thing, the idea that – given the right set of conditions – it can aggregate the intuitions of thousands and thousands of different traders across the entire globe to provide a glimpse into their collective insight. This… does not always work. Nevertheless, it can be somewhat less bad than alternative frameworks of trying to figure out what is happening. Even a deeply mistaken market will, eventually, update when enough of the chuckleheads have gone bankrupt.
Yes, but “the market” can still get the “right answer”, given certain conditions, even if every single individual out there buying and selling is rather grossly mistaken about why they’re trading.
I think I heard someone say the other day that inverted yield curves had predicted 18 of the last 6 recessions.
But a recession is itself just “a collectivization of a whole lot of individuals making decisions” – loss of consumer confidence leading to lower consumer spending, lower manufacturing investment, stock market sell-off – and bingo, suddenly you’re in a recession. IANAE, but this is why the inverted yield curve is considered the first harbinger of this cycle and therefore of some value as a predictor. It may not guarantee a recession is coming, but I think CNN was reporting that every single recession, at least in modern times, has been preceded by an inverted yield curve leading to recession within 6 to 24 months.
Really?
::goes to Schwab to check on bond yields::
By George, you’re right. 1 month Treasury bonds are paying as high as 1.9%, and everything longer term than that is paying less.
It’s a good place to stash a bit of cash for the short term.
In general, high interest rates are associated with a strong economy, and low interest rates with a weak one. This is due to a variety of related reasons, e.g. inflation, returns on competing forms of investment, etc. The latter is important in its own right, in that interest rates are also a proxy for expected returns for the market in general.
If the short term interest rates are higher than long term rates, then the collective wisdom of the markets is saying that market returns over the longer term are likely to be lower than returns over the short term. This indicates a belief that economic conditions will deteriorate in the future, as compared to what they are today.
Of course, that collective wisdom can be wrong. But it’s not something to dismiss lightly either. This is partly because the collective wisdom of people who, in aggregate, understand the market and have their own money at stake is worth a lot as compared to other forms of assessment (i.e. individuals spouting off various theories), and partly because collective belief that a recession is coming can be a self-fulfilling prophecy, as people and companies react by cutting back on investment and spending etc.
That said, one possible reason to think that the value of this indicator may be lower than history would indicate is that interest rates are subject to a high degree of manipulation by central bankers these days, making it less of a pure indicator of market sentiment than it might have been the case otherwise.
If institutions make the same decision, doesn’t that mean that the market will self-correct. If loads of people want the 1 month Treasury bonds, the return will go down.
It’s not an automatic indicator. It could point to structural issues like falling population growth or a fall in the currency. While US population growth is slowing down somewhat, the dollar is strong and rising, so it’s probably not that.
But the past few recessions have always been preceded by the inversion, so it definitely shouldn’t be shrugged off without taking a good look at the other signals.
No. The original terms don’t matter because it’s the secondary market that determines the price (and therefore present-day yield).
If you wanted to ask why the secondary market is doing what it does, you would just as institution managers or financial analysts. In other words, the same as the stock market.
I’m not an economic expert, but analysts are calling out other recession signals as well. Other safe investments like gold and US dollars are increasing. Industrial activity signals such as copper prices are down. They point to risks like the looming Brexit deadline in October with no resolution in sight. China is showing signs of softening activity, and Trump is making it worse with his dumb trade war.
tl;dr the yield inversion is not a distinct recession predictor, but a lot of other signals seem to be agreeing with it.
Keep in mind one other key point about the “wisdom of the crowds”.
It’s not simply the number of market participants. Most of the “votes” of the market (trades that result in the largest price changes) come from market players with the most money.
This is why long term the theory that the market is usually right has some validity. Because over time, market participants who make bad decisions will lose money and wise participants who make good decisions gain it, and then the richer participants are the majority of the “votes” that you are seeing.
It’s not joe daytrader inverting the yield curve. It’s massive institutions who have the resources and access to high quality information, qualified skilled traders, and these days, the state of the art in artificial intelligence.
Does this mean they are always right? No, but it is not completely unfounded to think they might be right.
Good question.
The weird part about the 1-month, tho, is that it is heavily influenced by Fed policy decisions. The 1-month is an extremely close substitute for federal funds or holding excess reserves, which mean the rates for these options always move strongly in sync. If the price of the 1-month is rising from more demand (i.e. the yield is falling), which is causing the fed funds rate also to rise outside the Fed’s target band for that rate, then the Fed can start unloading its own supply of 1-month bills to push the rate back up where they said they wanted it to be.
But if the yield curve has inverted, it’s very likely that the Fed will look at its current target band for interest rates, and say that that target is too high. They are likely to drop rates. So what you can say can absolutely happen, if it’s allowed to happen.
This is backward. It seems like maybe you have nonstandard definitions for long term and short term.
Think of it like this: I challenge you to predict the future. If you’re right, I pay you a million dollars, if you’re wrong, you pay me a million dollars. You can choose the timeframe you want to predict. What’s going to happen in 2 years, vs. what’s going to happen in 10 years. Which do you feel more confident predicting?
Today the market is telling us that the 2-year horizon seems less predictable than the 10-year horizon. So that is cause to take note.
Oh yeah. The market always corrects, over time. The problem is that you, bob++, have no way of knowing the definition of “correct”, or when it will happen.
It could be that the “correct” levels are even worse than they are now. Or maybe things will get better like they did last March. We don’t know what “correct” is or when it will happen.
No idea what this is about.
This is incorrect. The long term is always less predictable than the short term. The long term consists of the short term, which has its unpredictable aspect, plus the subsequent term which has its own unpredictable aspect. Unless you believe there is some negative correlation between short and long term, it’s not possible for the long term to be more predictable than the short term (on an absolute basis).
However, predictability (aka “risk”) is only one factor in interest rates. There is always a risk load to longer term rates. But another factor is market assessment of future conditions. If the market collectively believes that market conditions and returns will deteriorate in the future, then it could price the long term rates lower, despite the higher risk.
On that note, it’s worth noting that there’s nothing magical about an inverted yield curve, IOW it’s not all-or-nothing. Even a flattening of the yield curve attracts the attention of economists, because even a relatively flat (but not inverted) yield curve may also suggest negative market sentiment about the future - if not enough to overcome the risk premium. But inversion is an easily measurable and noticeable number and for this reason it attracts a lot of attention.
Depends which points on the yield curve you compare. But in cases where the 10 yr US govt bond yield is less than the 2 yr US govt bond yield, which is what made the stock market freak out earlier this week, a recession has followed within some months to a couple of years or so every time in recent history.
But that’s been a significant delay in many cases. And it’s not a causal relationship that must hold. This could be the time 2’s-10’s inversion isn’t followed by a recession even within a couple of years, and then everyone can reach for explanations why the relationship broke down, and there are candidate reasons.
More accurately, participants who make unlucky decisions will lose money and participants who make lucky decisions will gain money. If a way to make more money than others were predictable, the market would very quickly act such that it was no longer a way to make more money than others. And having more skill or wisdom is something that’s predictable.
To put it another way: If I thought that Warren Buffet, say, were better at picking investments than others (based on large quantities of information or wisdom of experience or just good gut instincts), then I could make a large quantity of money just by mirroring what Buffet does, even without having access to his information or wisdom or gut: If he buys stock in a particular company, I’ll buy proportionately to him in the same company, and if he sells, I’ll sell in proportion to him. And if he were actually consistently good at it, pretty soon, everyone else would be copying him, too, and now he’s no better than the market as a whole (though, if he’s genuinely good at it, he might have improved the market as a whole).
The only way to beat the market as a whole is with inconsistent good decisions, i.e., luck. And so there’s no correlation between who’s made the good decisions recently and who will make the good decisions in the future.