The Yield Curve Posted by: Dale Franks
on Thursday, January 12, 2006
Is another recession in the offing? Those who watch the yield curve carefully are getting concerned. As of right now, the yield curve is inverted. Pretty scary, no?
Now, of course, many of you are scratching your heads at this point and asking, "What the hell is a yield curve?" and "Why do I care if it's inverted?"
Aha! Time for an economics lesson!
First, let's talk about yield. The government sells a variety of treasury bonds. Some of the bonds are short-term bonds of 1 year or less. Some of them are long-term bonds of 30 years. Each of these bonds pays interest of a certain percentage. For instance, when you buy a 30-year bond, you'll slap down a minimum on $10,000. In return, for your 10 grand, the bond will pay what is called a coupon rate of, say 7%. That means that, if you buy a $10,000 "thirty", then the government will pay you $700 per year for 30 years, then they'll give you back your initial 10k. This $700 per year is what is called the coupon payment (because, back in olden times, bonds came with coupons that you'd have to mail into the Treasury Department to get paid). The coupon payment never changes. Whoever holds that bond will get $700 per year. Not a penny more. Not a penny less.
Unfortunately, most of us, since we don't own a bank, can't buy a 10,000 bond when it's issued by the government. Instead, we can only buy a bond from someone who already owns one. This is called buying a bond in the secondary market. Now, when you buy a bond in the secondary market, the bond has to pay you an interest rate that's equivalent to the rate of return you can get on other investments. If it doesn't, then you'd buy something else, of course.
But a bond only pays a set amount of interest, based on the face value of the bond. So, if a bond's coupon payment is only $700 bucks a year, and other investments are paying 10%, I can't sell it to you for $10,000, because you'd only be making 7% on the bond. I have to lower the price of the bond so that your $700 coupon payment is the equivalent of 10%. If I sell you the bond for $7,000, then that $700 coupon payment becomes a 10% yield. So, the yield is the rate of return you get on a bond, based on the price you paid for it.
By the way, just so you know, if interest rates are at 7%, then the price of the bond in the secondary market would be the same as the face value of the bond, $10,000. This is known as buying the bond "at par", i.e., the price in the secondary market is the same as the face value of the bond. If you pay less than $10,000, then you've bought the bond "below par". If you pay more than $10,000, you are buying the bond "above par".
So, now we know what the yield is, let's talk about the yield curve. As I mentioned before, the government sells bonds that mature in different amounts of time. There are Treasury Bills that pay you back in 1 month, and the longest term bond matures in 30 years. So, to create a yield curve, you line up all the bonds from the shortest maturity to the longest maturity, and plot the yields each bond is paying. Normally, the shorter the term of the bond, the lower the yield the bond pays. So, in normal circumstances, a 1-month Treasury note wil pay a yield of 5.75%, and interest rates will increase all the way to the 30-year bond which would pay 7%. So, a normal yield curve might look like this:
Looking at a normal yield curve, you can see that short-term rates have a lower yield than long-term rates. This is because bonds trade on expectations of inflation. So, the longer the term of the bond, the greater the danger of inflation cropping up before the bond matures. So, you see a gently sloping yield curve, with rates rising as the term rises.
But, at the moment, and since 27 December, the yield curve hasn't looked like that. It's looked like this:
As of today, the yield for the 6-month bill and the 10-year note have been about the same, while yields between 6 months and 10 years have been lower. When short-term yields are higher than long-term yields, the yield curve is said to be inverted.
This is taken by a lot of people to be a bad sign. Quite often, an inverted yield curve is a sign of a coming recession.
But, despite the inverted yield curve, the other fundamentals of the economy look strong. Economic growth has been 4%+ for the last few quarters. Employment growth not only remains strong, but employers are forecasting strong hiring for the next year. So, what does this inverted yield curve mean? Should we be scared?
A couple of factors come into play hear. First, the Fed has been reining short-term interest rates for a while now. When the Fed increases the Fed Funds Rate and the Discount Rate, both of which are very short-term rates, then short-term treasury rates tend to rise in response. Long-term rates rise, too, but not as much as short-term rates.
In the case of the Fed's current rate-increasing environment, an odd thing is that, while long-term rates have risen, they've risen much more slowly than they have in the past. Why could this be?
While various factors affect long-term rates, economists generally see them as an average of current short-term rates and the short-term rates traders expect in the future, says Nicholas S. Souleles, finance professor at Wharton. Current yields are known, but long-term yields can only be guessed at. They largely depend on what the Fed will do with short-term rates in the future, and that is governed not just by evolving Fed philosophy but by the unpredictable factors it will evaluate years down the road.
Simple arithmetic says that if the Fed lifts short-term yields, long-term ones will follow — but not by as much, since current short-term yields are only part of what governs long-term yields, Souleles says. "We always see this: When the Fed raises short-term rates, long-term rates don't go up as much." Typically, long-term rates rise about two-fifths as much as short-term rates do.
But today, "long-term rates have not gone up as much as they would typically," he says, adding that economists have focused on three general reasons for this. First, bond traders may be anticipating low inflation in the future, which would allow the Fed to keep interest rates low, or to make them even lower. Some economists and traders believe that globalization will rein inflation in, as more products and services are produced by low-cost economies.
Also, he adds, the bond market may be anticipating an economic slowdown or recession as well as a Fed rate cut to make more money available to stimulate the economy.
Second, long-term rates may be staying low because high demand for Treasuries and other U.S. debt securities keeps bond prices high, which keeps yields low. Bonds represent loans from bond buyers to bond issuers. When demand is high, issuers like the government can attract lots of buyers despite offering low yields.
Various factors affect demand for Treasuries, including their perfect safety record. But demand has been increasing, Souleles says, because China, Japan and some other countries are selling more products to the U.S. than they are buying, leaving them with a cash surplus they are stashing in safe Treasuries. "The Chinese have more income than they are spending. They are saving those funds and some of the saving is going abroad."
The third reason for low long-term yields involves traders' demand for a "risk premium," according to Souleles. Typically, they demand higher yields to offset risks related to tying money up in long-term bonds. For example, if interest rates rise in the future, bonds issued at that time will be more generous than ones issued today, so there will be less demand for the older bonds and their prices will fall. Similarly, higher inflation in the future could chew away a good part of a bond's interest earnings.
These risks are not as pronounced for short-term bonds, since they will soon automatically convert to cash that can be reinvested in whatever way seems most suitable for the changing conditions. So there is little risk premium on short-term bonds.
When long-term rates are virtually the same as short-term ones, it could mean that traders don't believe it likely that interest rates and inflation will move higher, Souleles says. Hence, they do not demand as high a risk premium as they did in the past.
Early in January, the Fed released minutes of its December meeting, indicating that its rate-raising cycle may soon come to an end. If short-term rates will not be rising in the future, that would help keep long-term rates low.
This is an interesting phenomenon. It may be that, because of globalization, long-term rates now have pressure to stay lower than they otherwise would. If so, then looking at the yield curve and seeing an inversion may not mean that a recession is in the offing, as it often has in the past. We may simply have to look at long-term rates in a different way.
Given all the factors that can be affecting today's yield curve, a looming recession can hardly be considered a certainty. But that cannot be entirely ruled out, either, notes Francis X. Diebold, professor of economics, finance and statistics at Wharton.
In fact, most recessions have been preceded by inverted yield curves, as traders anticipate a Fed rate reduction to stimulate the economy, he points out. "I'm not going around saying there's a recession coming." But he notes that the future is not necessarily rosy, either. "The bond market doesn't seem to be as worried about inflation as I am," he adds, arguing that growing demand for oil by China, India and other countries will continue to put upward pressure on energy prices, contributing to broader inflation.
Interest rates are also likely to rise as the U.S. deals with the huge federal budget deficit, according to Diebold. There are only three ways to address that problem — raising taxes, borrowing or printing money. The Bush administration has ruled out higher taxes, and the other two remedies both tend to drive interest rates up.
Hence, he predicts inflation will run 3.5 to 4% over the next decade, a half to 1 percentage point above the long-term average. Investors in 10-year Treasuries will demand a real return of about 1.5 points above inflation, and they will want a 1-point risk premium. That would take the 10-year Treasury to 6 or 6.5%, well above today's 4.3%.
And that would be the end of today's partially inverted yield curve. "Markets are fickle," Diebold says. "So I can't say this with any certainty, but I wouldn't be surprised to see the yield curve steepening, with the long bond going up."
So, if inflation is on the rise due to rising energy demand which increases the price of energy supplies, and eventually translates into higher prices in general, the implication isn't good. Higher inflation means that the Fed will have to implement a tighter monetary policy, which means higher rates in general, and slower economic performance.
Because, there is a line of economic thought that holds the opinion that a little inflation serves as grease to the wheels of the economy. There is, of course, another line of thinking that says the pro-inflation guys are nuts.
So, what is happening? Will globalization depress long-term rates, eliminating the effectiveness of yield curve inversions as a predictor of recession? I expect we'll find out over the next three quarters.
If the 10k-30 bond is paying me 700 a year for 30 years, this is simple, not compounded interest, correct? Is that how federal bonds work? Because 10k invested over 30 years with a 7% annually compounded rate would net me about 81k in the end, but as you describe it I’d only have 700 x 30 + 10k or about 31k.
Yes, Mr. Franks’s price-to-yield computation is oversimplified. In his example, if I buy a bond with 7% coupon for 70% of notional, the coupons give me a 10% yield, but I also get 100% of notional back at maturity — so the yield is greater than 10% (much greater for short-dated bonds).
Approximate yield-to-price computations aren’t too hard, if you know how to collapse a geometric series; but price-to-yield is always done numerically and doesn’t really lend itself to a one-paragraph explanation.
I don’t think this undermines the substance of Mr. Franks’s article.
Given all the factors that can be affecting today’s yield curve, a looming recession can hardly be considered a certainty. But that cannot be entirely ruled out, either, notes Francis X. Diebold, professor of economics, finance and statistics at Wharton.
It might rain today. Or it might be sunny. Or maybe clouds and a few sprinkles.
I seem to recall the last time the yield curve inverted was early 2000. (I don’t know whether you mentioned this, Dale, I didn’t see it in a quick reading of your post.) Of course, past results are no guarantee of future performance.
What does it mean? (yoda) Difficult to say. Always changing, the future is.(/yoda)
Inflation is an alteration of the supply of and/or the demand for money such that its perceived purchasing power falls. Normally, it is the artificial expansion of the money supply that causes this alteration. Whether prices rise or not depends upon real supply and demand factors for the goods and services in question. Rising prices for oil due to alterations in the supply of and/or the demand for oil cannot and do not imply inflation (ie Katrina, Rita, or the uncertainties associated with the politics of Saudi Arabia, Venezuela, Mexico, Nigeria, Iraq, Iran, etc).