In the endless debates over stimulus and deficits, more “dovish” commentators frequently point out that debt markets appear sanguine about US borrowing. Despite some recent upward jitters, the Federal Government currently pays less than 4% to borrow for 10 years, and under 5% to borrow for 30 years. Those are bargain rates in historical terms, the argument goes, so investors must not be terribly concerned about inflation or default or any other bogeyman of “deficit terrorists”.
To make the point, Paul Krugman recently published a graph very similar to this one:
Since the financial crisis began, the US government’s cost of long-term borrowing has dramatically fallen, not risen. If we graph a longer series of 10-year Treasury yields, the case looks even more compelling. The United States government can borrow very, very cheaply relative to its historical experience.
However, there is another way to think about those rates. The US government’s cost of long-term borrowing can be decomposed into a short-term rate plus a term premium which investors demand to cover the interest-rate and inflation risks of holding long-term bonds. The short-term rate is substantially a function of monetary policy: the Federal Reserve sets an overnight rate that very short-term Treasury rates must generally follow. Since the Federal Reserve has reduced its policy rate to historic lows, the short-term anchor of Treasury borrowing costs has mechanically fallen. But this drop is a function of monetary policy only. It tells us nothing about the market’s concern or lack thereof with the risks of holding Treasuries.
But the term premium (or “steepness of the yield curve”) is a market outcome (except while the Fed is engaged in “quantitative easing”). How do things look when we graph the term premium since the crisis began?
The graph below shows the conventional barometer of the term premium, the 2-year / 10-year spread (blue), and a longer measure, the spread between the yield on 3-month T-bills and 30-year Treasury bonds (red), since the beginning of the financial crisis:
Since the financial crisis began, the market determined part of the Treasury’s cost of borrowing has steadily risen, except for a brief, sharp flight to safety around the fall of 2008. Investors have been demanding greater compensation for bearing interest rate and inflation risk, but that has been masked by the monetary-policy induced drop in short-term rates.
Taking a longer view, we can see that the current term premium is at, but has not exceeded, a historical extreme:
Note: There’s a gap in the 30-year rate series, probably because it became impossible to compute a “constant maturity” 30-year Treasury yield during the period when the Treasury stopped issuing 30-year bonds.
The present term premium is quite similar to those that vexed President Clinton during the heyday of the “bond vigilantes”. A glass half full story says that, despite all the stresses of the financial crisis and the sharp spike in Treasury issuance, the term premium has not become unmoored from its historical range. A glass half empty story says that the term premium is toying with the boundary of that range, and could break loose in an instant. I have no idea which tale is truer.
My politics on the deficit are centrist to dovish. I think that deficit spending is always an option, and that the Federal Government should absolutely spend on forward-looking, high-return investment projects. I also favor generous “safety net” benefits and would like to see a guaranteed income program in the United States. However, the only form of “stimulus” I support is very broad based transfers (e.g. I would support a payroll tax subsidy). I agree with many left-ish commentators that the deficits we’ve experienced are more an effect than a cause of our economic problems, and will take care of themselves if we create a strong economy and a broadly legitimate political system (neither of which I think we have right now). A nation as large and wealthy in natural resources and human capital as the United States need never be constrained by the vicissitudes of financial markets, if its government is capable of mobilizing its citizens’ risk-bearing capacity on behalf of the polity. [*]
But whatever my politics, I think there is a fair probability that the US will experience the thrilling uncertainty that attends a loss of confidence in its currency and debt. The argument that “markets don’t seem troubled by our deficits” is less persuasive than it first appears.
Full disclosure: Although my views are sincerely held, sincerity is cheap. Perhaps I am just talking my book! I am one of those people long gold and short Treasuries (although I am not a proponent of the gold standard).
[*] Remember, economic capital has nothing to do with money. Supplying capital is nothing more or less than assuming the burden of economic risks. Where do you think China’s ever-expanding capital base comes from, when it has been the world’s largest exporter of financial capital? China’s citizens assume great risk, in the form of below-world-market wages and social safety benefits, in exchange for the promise of a wealthier and more powerful nation. To some degree that capital is extracted involuntarily, but China’s government has had remarkable success at maintaining legitimacy and the consent of the governed despite the extraordinary costs citizens have borne in the service of an uncertain future. So far, citizens have seen consistent returns on their investment: the big question is how China fares in a persistent “bear market”, when it comes to seem as though much of their sacrifice has been wasted or stolen.
Hi Steve,
For some reason, the pictures aren’t showing. When I copy the URL (e.g., http://www.interfluidity.com/v2/files/yield-10yr-crisis-period-2010-03-28.png) and paste it in a separate window, I get a “page not found” error message.
Michael
SRW,
1. The yield curve doesn't reflect risk alone. It reflects expectation, which is not risk per se. We should differentiate between the expectation for future short rates and the risk around that expectation.
2. The current yield curve reminds me of Greenspan's "conundrum". He was puzzled that long rates didn’t increase when he tightened fed funds last decade. It's part of his explanation as to why he blames the "abnormal" non-tightening response of long rates for the housing bubble, rather than his low short rate period per se. I never believed that a conundrum existed, and I see it more or less repeating that way (as an observation, not a central bank complaint) in the next up cycle "“ i.e. the expectation for required tightening is reflected mostly in today's steep curve. The initial steeper curve in both cases reflects the structural uniqueness of an immovable zero lower bound. It's a coiled spring yield curve as opposed to one that floats up or down by movements at both end. (A classic counter example is the 1993-1994 tightening cycle, which moved long rates substantially on a proportionate scale, starting from short rates at around 3 per cent). I don't see the curve as a big risk premium. I see it largely as the expectation/hope of more normal yield levels and a curve that trends to flatter when the Fed tightens.
3. Short treasuries may work as a trade, more in 2's than 10's, but get out when resulting curve triggered re-deflation risk extends the expectation for low Fed policy rates.
Pictures show with explorer but not with chrome.
Michael, JKH — Thanks for letting me know about the broken images. I was using relative links which resolved (as Michael found) to a bad URL, but unfortunately Safari and apparently IE are too clever by half and covered up for my mistake, so I didn’t notice…
JKH — Yeah… I know that usually people subdivide what I’m calling the term premium into the expected interest path and the term premium due to liquidity/unexpected inflation(/default?) risk. I’ve agglomerated all of that into an “all risk” term premium. From the perspective of a long-bond buyer, an interest rate rise is a risk for which she demands compensation, even if it is priced into forward yield curves. (The change is not certain, even if it is in some sense “the market’s” expected value.) From the perspective of evaluating market evaluations of the “trustworthiness” of a government with respect to its currency management, the expected interest rate component has to be in the measure, because market participants may expect interest rates to rise due to an uptick in real activity (triggering an ordinary monetary policy response), or due to inflationary pressures (triggering either a monetary policy response or an inflationary accommodation, in either case requiring compensation). If markets are anticipating inflationary pressure (whatever the response), that’s substantively a point against the “bond yields prove that markets are unconcerned” view.
One benign explanation for this “conundrum” is that there is a “normal” long-term rate, and extraordinary low short-term rates have just pulled down the spread relative to a fixed ceiling. (From some theoretical perspectives, very long-term interest rates should literally be unchanging.) That may be right, but it is not obviously vindicated by the data: in practice, US long-term rates have been quite variable, and one would characterize them more as trending than mean-reverting. I think this is similar to your “coiled spring” view.
I won’t argue that these more benign interpretations are wrong. It may well be that markets are anticipating a solid recovery and a normalization of monetary policy, which might well lead to a Greenspan-conundrum-like flattening of the yield curve, with short rates rising and long rates falling as whatever part of the long end is risk-related falls away. The consistently moderate spread between TIPS and unprotected Treasuries argues for a not-inflationary interpretation (although even in theory that can’t differentiate between not-inflationary because the Fed fights back at the expense of output from not-inflationary because the economy recovers). Overall, I don’t think that bond markets (or markets generally) arbitrage well over long periods of time, though, so I’m skeptical of reading very much into the TIPS/Treasury spread. It’s a data point worth considering, but I think clientele effects as well as the widely accepted liquidity effects affect the price of TIPS more than they hypothetical long-term arb.
The elevated term premium / steepness of the yield curve is certainly not a “slam dunk” case for market skepticism of the Treasury as a borrower. But neither are the low absolute yields a strong case for market confidence. Overall, I think don’t think yields tell us very much either way, they just become a market Rorschach test in which commentators see whatever they want/expect to see.
p.s. in the little China postscript, there are shades of your influence, specifically your long-ago conjecture that taxation is analogous to equity in government. i think it is very interesting to consider ways in which people provide capital, bearing economic risk in exchange for uncertain future benefits but without formal assets to mark their claims. consensual taxation (meaning despite grumbling, taxation widely perceived as legitimate with which individuals generally comply) is very much like equity — you part with your money in hopes that it will be used well, and generate public goods worth the personal cost. of course, nonconsensual expropriation (of money taxes or labor) can also generate capital, it is just less analogous to market equity. i think China’s capital conundrum — a recently capital-poor nation that has continually expanded both its internal capital base while exporting financial capital — is fascinating, and helps think about the deeper sources of economic capital.
When short-term rates are artificially low, it seems that the term premium is less useful as an indicator. As the fed moderates short-term rates, long rates won’t necessarily rise an equivalent amount, and this would reduce the term premium.
[...] This post was mentioned on Twitter by Edward Harrison, Abnormal Returns Now. Abnormal Returns Now said: interfluidity » A different perspective on interest rates http://stk.ly/9bTWXv $$ [...]
[...] What happens if markets get spooked by US government deficits? (Interfluidity) [...]
A little off topic, but not too much since it still involves those bond interest rates. Perhaps you’ve seen the whole Kinsley-Krugman (hyper) inflation pas de deux. Much ado over nothing, I surmised, since they did much to talk past each other. Krugman tries to distinguish between “moderate” inflation (say, 4 to 20%) and “hyper” inflation (some magical cutoff number that includes Zimbabwe and the Weimar Republic?).
Read Full Article »