April 04, 2006
What's moving long-term yields?
Long-term interest rates continue to creep up.
The interest rate on 10-year Treasuries is now 50 basis points higher than it stood at the start of the year. This rise coincided with a run-up in many commodity prices, leading some such as Capital Spectator to speculate that the Fed is losing the fight to persuade markets it can control inflation.
I commented earlier that the twenty-basis-point increase in ten-year yields in January and February did indeed seem to reflect concerns about rising inflation. However, the chart below updates those data on the 10-year nominal yield and inflation-insured TIPS yield. Since both the capital value and coupon of the latter are guaranteed to rise with the consumer price index, movements in TIPS yields cannot be due to inflation fears. Yet clearly nominal and inflation-adjusted yields have moved very much in tandem over the last month. The 30 basis point gain in long-term yields over that month certainly looks to me like a rise in real yields, not inflation concerns.
It makes sense to look for a common factor behind the commodity bull market and the fixed-income bear market. What might that common factor be, if it indeed is not a concern about inflation? One possibility is that the market has revised its expectation of real economic growth upward. Stronger real growth would be a factor that would boost demand for both commodities and investment borrowing.
If so, is the market right about that forecast? A very strong value for first-quarter GDP growth is certainly something we might look for at this point. As for the rest of 2006, I'm not ready to jump on that bandwagon just yet.
Posted by James Hamilton at April 4, 2006 09:30 PMdigg this | reddit
Listed below are links to weblogs that reference What's moving long-term yields?:
» Mohamed's US dollar from Contango
“If Mohamed will not go to the mountain, the mountain must come to Mohamed” If the USD does not depreciate, then all prices must rise. The Chinese have not allowed their Yuan to appreciate; on the other end, the Fed [Read More]
Tracked on April 6, 2006 01:56 PM
I'm on the same wavelength. I recall that commodity prices are recovering from a prior slump, which might simply reflect increased global demand. Modern economies use less physical resources per unit of GDP and the impact of rising prices is not as significant as it was say 30 years ago.
I used to think that the gold price was a good indicator of inflation, but I'm not so sure now. Since gold peaked around $800 per ounce around 1980, it certainly has not kept its value after inflation. I also understand that the gold market is relatively small compared to international currency flows and can be influenced by relatively small changes in demand.
Posted by: Rich Berger at April 5, 2006 05:13 AM
Another possibility is that the market is concerned about incipient inflation but has confidence (at least no less than a month ago) that the Fed will succeed in eventually getting that inflation under control. The argument is that the Fed impacts the economy primarily through its effect on long-term real interest rates, so that the Fed will somehow have to get long-term real interest rates to rise in order to slow down the economy sufficiently to offset the incipient inflation. And of course, TIPS traders know better than to fight the Fed.
Also, I think there is widespread opinion that it is foreign growth, rather than US growth, that is at issue. If Japanese and European growth is more than originally expected, their interest rates will rise and lead people to shift assets out of US Treasuries. (I have argued in previous comments in this blog and elsewhere that this type of shift need not ultimately affect US interest rates unless it reduces the trade deficit, because the dollars that go abroad in trade will still have to be invested in the US. However, it is also reasonable to expect that foreign growth, and the dollar weakness that goes with it, would reduce the trade deficit.)
Posted by: knzn at April 5, 2006 05:29 AM
Over the past couple of years a few culprits have been identified as causing low long-term yields including foreign central banks managing their currencies, China & Japan, oil exporters not spending their windfall, Saudi Arabia & Russia, and pension funds moving out of stocks and into bonds to better match their long term liabilities.
With China slowly moving to a floating exchange regime, Japan ending its quantitative easing, oil prices stabilizing/declining, and pension funds possibly completing their asset reallocation, current long bonds may be reflecting a drying up of demand.
Posted by: kashof at April 5, 2006 06:59 AM
I think kashof has it right. Bond buyers often are not particularly prescient regarding future inflation. I think you are an idiot to lend anyone money for 30 years for below 5%.
Posted by: algernon at April 5, 2006 07:29 AM
Check your premises.
The market probably knows that the government's CPI number is low-balled. Roll back the changes to CPI made in the 90s, and you get a number 3-4% higher.
If this is the case, then even TIPS are still exposed to *real* inflation.
Posted by: Aaron Krowne at April 5, 2006 08:06 AM
U.S. economic growth is of lesser relevance for commodity prices. More important is growth in China and other emerging economies where economic growth is more natural resource intensive.
Posted by: Stefan Karlsson at April 5, 2006 08:49 AM
Regarding Aaron Krowne's point about low-balled CPI numbers, John Williams of Gillespie Research tracks what he calls pre-Clinton Era CPI (http://www.gillespieresearch.com/cgi-bin/bgn/). His pre-Clinton Era CPI figure stands around 6.5%. So, are Treasury yields providing positive or negative real yields? If the latter, why hold them?
Also, foreign central banks may finally be reducing their ongoing demand for dollar reserves. Naturally, a reduction in dollar reserves isn't necessary for Treasury prices to go down, but rather a mere reduction in purchases of Treasuries. Having said that I seem to recall that, in March, Xiao Zhuoji, a member of the Chinese People's Political Consultative Conference, said that most of the People's Bank's forex reserves are dollar-denominated and that its forex reserves should be cut by 2/3.
Posted by: Suresh at April 5, 2006 09:27 AM
Could it be that it is just the tendency to revert to some 'fair value'? Currently, I estimate with a simple cointegration model, which includes a real activity measure, inflation and a short term rate, an equilibrium of 5.6% for 10-year Treasuries. Any fundamental model I've seen points in this direction.
Posted by: Anonymous at April 5, 2006 09:48 AM
Changes in central bank reserves, I would argue, don't matter for US rates except to the extent that (1) they affect the trade deficit or (2) central banks have different asset class preferences (bonds vs. stocks, etc.) than the private sector. If central banks reduce their dollar reserves, they have to sell dollars. This creates a surplus of dollars. Eventually, these dollars will have to be offered cheaply enough that someone is willing to buy them, and whoever buys them will have to invest them in US assets. Of course, the private sector may well have different asset-class preferences than central banks, and reserve sales may well affect the trade deficit, but the major symptom of reserve sales will be a drop in the dollar, not an increase in US yields.
Posted by: knzn at April 5, 2006 10:06 AM
Aaron, your argument has to be not just that the CPI is an imperfect measure of inflation, but further that it has nothing at all to do with inflation. Surely, if inflation goes up, there should be some increase in the CPI, even if the relation, in your view, is not one-for-one. Therefore, if inflation expectations go up, there should be some increase (perhaps not one-for-one) in nominal yields relative to TIPS. But there's no evidence of any movement in the spread at all over the last month, which leads me to conclude that a change in expectations of inflation cannot be the factor that accounts for the rising nominal yields.
Stefan and others, I agree foreign growth is likely more important than U.S. growth, though I expect the two are related. I point to U.S. 2006:Q1 GDP in part just because that's one statistic I've come to be more optimistic about, and so it's natural for me to think that the market has been doing the same.
Posted by: JDH at April 5, 2006 10:13 AM
Is it possible that TIPS is rising because investors feel the inflation measures used aren't revealing the full situation of inflation, ie some inflation levels are hidden?
Posted by: jldugger at April 5, 2006 10:17 AM
In AFP's article entitled, "China's forex reserves much too big, offer US cheap financing," Xiao Zhuoji (an economist by trade) said that most of China's foreign exchange reserves are mainly invested in low yielding US treasuries (government bonds), effectively providing "low-cost" financing for Washington. http://22.214.171.124/search?q=cache:EORlyCtF5-kJ:news.yahoo.com/s/afp/20060303/bs_afp/chinaforexyuan_060303055128%3B_ylt%3DA86.I0sZsx1EDA4AOREMO7gF%3B_ylu%3DX3oDMTA5bGVna3NhBHNlYwNzc3JlbA--+China%27s+forex+reserves+much+too+big,+offer+US+cheap+financing+&hl=en&gl=us&ct=clnk&cd=1&client=firefox-a
Posted by: Suresh at April 5, 2006 11:44 AM
Anything that China might be doing to cause the rise in US interest rates, they are not actually doing yet, but it might be the market's anticipation of such action that is causing rates to rise. If China were selling off dollar-denominated assets, we would be seeing the effect in a weaker dollar. But Zhao also said that China needs to "encourage capital outflows and should allow companies and individuals to hold more foreign currency." Such actions would not weaken the dollar, but we would presumably hear about them directly. If the market is anticipating that China will take actions to encourage its private sector to buy US assets, and if the market anticipates that those assets are not likely to be bonds, then this could partly explain the rise in yields. As the Chinese private sector buys non-bond US assets, the central bank can sell US bonds (or, more precisely, stop buying quite so many) without affecting the exchange rate.
Posted by: knzn at April 5, 2006 02:27 PM
Imagine that you are managing pension fund money, or running an insurance company. You have long liabilities to worry about, and the usual response is to buy long assets, in order to match. But with 30 day rates as good as 30 year rates, and all the indicators including wage costs and commodities pointing to higher inflation, maybe you would choose to create a mismatch by buying short assets. Even more so if you can roll over your money at an added 25 bps in 60 days or so. Multiply this by all the other people thinking the same way and it is not hard to imagine demand for long assets drying up at the margin. Pretty soon you have long rates moving up 50 to 100 bps. Isn't that how markets are supposed to work?
Posted by: David Baskin at April 5, 2006 06:46 PM
It seems to me that an expectation by foreign investors that the dollar will decine in relative value would explain rising interest rates and , partly, dollar valued commodity prices. Holding a dollar denominated bond while the dollar declines relative to your home currency will result in a capital loss and tend to increase the expected interest rate to entice an investor to lend.
Likewise, the dollar price of commodities will tend to rise vis a vis the price denominated in other currencies.
Although commodity prices have risen far above the levels expected by a simple currency ratio. There seems to be a real, or otherwise, belief in long term, above average, synchronis world growth in the immediate future and an expectation of shortages in all commodity classes. This belief seems to be an outgrowth that there is not enough to go around, exaserbated by a deficit in development of new production from the late 80's to the present.
Posted by: goof at April 5, 2006 08:18 PM
Bond yields going up represents weakness in demand for bonds. Lower demand means bond sellers have to offer higher yields to attract buyers. So we have weakness in demand for bonds coupled with strength in demand for commodities. Maybe investors are moving money from bonds into commodities to take advantage of the production squeeze going on there.
Posted by: Hal at April 5, 2006 09:59 PM
goof, In the case of freely floating counter-currencies, it seems to me that an expectation of a weaker dollar would result mostly in an immediate fall in the dollar rather than a rise in interest rates (though it could also involve some of the latter), unless bond market players have different expectations than forex players. (I've always been suspicious of arguments that different markets have different expectations. Why, when they have access to the same information?) In the case of RMB, or other currencies whose values are determined by intervention, there would be an increased demand to convert away from dollars. To meet this demand, China (for example) would have to increase its reserve accumulation, and thus probably increase its demand for US bonds in a way that would tend to offset the decrease in demand by the private sector.
Posted by: knzn at April 6, 2006 06:29 AM
Could it be that investors are losing confidence in the USD?
The BoC Australian administrator declared a week ago +/- that he would recommend apportioning a % of the BoC reserves in gold; which made the gold market gap and reverse to resume its bull direction.
Could it be that BoJ's end of its monetary easing is a landmark event?
BoJ's ending its monetary easing has investors looking at currency weakness from a different perspective, and unfortunately for the US --it will make the US CA deficit stick out to weight heavily on the USD depreciation, as Bill Gross indicates in his article:
Posted by: Joe Rotger at April 6, 2006 02:24 PM
Gold has risen precisely because real interest rates have been relatively low. The boost to the economy from the low interest rates is catching up faster than the FOMC.
Good call, Doctor Jim, real interest rates are rising. Speculators may take gold higher but real rates are are finally starting to make the "Gold Carry" expensive.
The world economy is so strong that the quarter point moves by the FOMC will not slow this train for at least a couple more quarters.
Posted by: Jack Miller at April 6, 2006 06:02 PM
"Bond buyers often are not particularly prescient regarding future inflation"
I don't think that could be any more wrong..that is pretty much the most important thing bond buyers look at...if your point is whether or not they are any good at predicting it, then you may be right, but nobody pays more attention to inflation issue more than bond folks.
I think that the re-awakening of Japan could have huge consequences...just the mention of it in the bond markets today sent the curve steepening like mad..and doubled the losses on the 10yr even beyond the Non-farm number this morning.
Fact is, if there is a demand shift out of treasuries due to more attractive yields elewhere, then treasuries will take a hit. The bid in the back end of the curve from non-economic buyers is what has brought us an artificially flat curve by about 50-60Bps...reverse that artificial flatness and we get back to normal (whatever the hell that is).
If Europe and Japan keep growing and increasing rates, I predict we will have a Bull market steepener on the way down as our economy cools. Why not right?
Posted by: Hit The Bid at April 7, 2006 01:39 PM