Sunday, April 11, 2010
On how malleable "economic" Theories can be
Back in business school, I was told that the spread of any interest rate over the "risk-free" short term government bill rate is a function of maturity risk, liquidity risk and default risk among other things. So, the higher the spread of the longer term Government bond rate, the greater the perceived uncertainty of the government's finances in the long run. Let's call this Theory A.
Theory B: At the same school, I was also told that longer term interest rates can be interpreted as the function of the market's expectations of short term interest rates in the future. For instance, one can think of the two year bond yield as the product of the current short term Fed rate and the market's expectation of the short term Fed rate one year hence.
I didn't note the contradiction between these two points of view back then. It all seemed very sober and scientific in the classroom. But now, I often wonder if these were political theories after all, masquerading as science.
The recent blog debate starring Megan McArdle of The Atlantic and Paul Krugman is particularly interesting. Here is the chart that spurred the debate (courtesy Steve Waldman). It shows that the spread of the 30 year/10 year bond yield over the 3 month "risk-free" rate has widened since the recession. Also, the spread is not showing signs of decline even after the recession subsided mid last year.
Megan is worried about this and interprets the trend as a sign of the market pricing in the perceived default risk posed by the US government. Like a good conservative Republican, she thinks this ought to be a warning to the government to stem the deficit and stop adding to the burgeoning public debt. I guess it is very clear that she subscribes strongly to Theory A.
Paul Krugman of the NY Times responded in typical fashion. Quite predictably, he said that Megan was all confused about yield curve basics. The long term bond yield remains high because the market expects short term interest rates to go up in future. Now, we all know that short term rates generally go up only if the Fed perceives improved economic times ahead. So Krugman is in effect saying - Don't worry about the yield spread. It is mostly good news!
Now, this is the classic Market expectations theory which we labeled Theory B earlier in the post.
The question is - which theory holds water? Krugman makes his theory sound more plausible, which is no surprise given his considerable verbal skills. But is he right? Can we be altogether sure that the long term rate is purely a function of future market expectations? If that's true, what about the default risk premia, liquidity risk premia and maturity risk premia that I read so much about in FM-1?
I'd like to believe that all these "theories" are political beliefs in the main. A liberal like Krugman who is so sceptical of Fama's efficient market theory and Lucas' rational expectations hypothesis is so willing to buy the prowess of the same inefficient market in predicting future Fed actions! It is convenient because the "market expectations" interpretation of the yield curve helps him shrug off critics who cavil at government spending and the huge debt burden by saying "Look...the bond market isn't particularly worried about the debt. Why bother!"
Similarly, the conservative Megan does not buy Theory B in the context of government bond market, but is quite likely to buy the not too dissimilar rational expectations theory which helps her make a case against expansionary government policy.
It's all so very politically convenient.
|
Back in business school, I was told that the spread of any interest rate over the "risk-free" short term government bill rate is a function of maturity risk, liquidity risk and default risk among other things. So, the higher the spread of the longer term Government bond rate, the greater the perceived uncertainty of the government's finances in the long run. Let's call this Theory A.
Theory B: At the same school, I was also told that longer term interest rates can be interpreted as the function of the market's expectations of short term interest rates in the future. For instance, one can think of the two year bond yield as the product of the current short term Fed rate and the market's expectation of the short term Fed rate one year hence.
I didn't note the contradiction between these two points of view back then. It all seemed very sober and scientific in the classroom. But now, I often wonder if these were political theories after all, masquerading as science.
The recent blog debate starring Megan McArdle of The Atlantic and Paul Krugman is particularly interesting. Here is the chart that spurred the debate (courtesy Steve Waldman). It shows that the spread of the 30 year/10 year bond yield over the 3 month "risk-free" rate has widened since the recession. Also, the spread is not showing signs of decline even after the recession subsided mid last year.
Megan is worried about this and interprets the trend as a sign of the market pricing in the perceived default risk posed by the US government. Like a good conservative Republican, she thinks this ought to be a warning to the government to stem the deficit and stop adding to the burgeoning public debt. I guess it is very clear that she subscribes strongly to Theory A.
Paul Krugman of the NY Times responded in typical fashion. Quite predictably, he said that Megan was all confused about yield curve basics. The long term bond yield remains high because the market expects short term interest rates to go up in future. Now, we all know that short term rates generally go up only if the Fed perceives improved economic times ahead. So Krugman is in effect saying - Don't worry about the yield spread. It is mostly good news!
Now, this is the classic Market expectations theory which we labeled Theory B earlier in the post.
The question is - which theory holds water? Krugman makes his theory sound more plausible, which is no surprise given his considerable verbal skills. But is he right? Can we be altogether sure that the long term rate is purely a function of future market expectations? If that's true, what about the default risk premia, liquidity risk premia and maturity risk premia that I read so much about in FM-1?
I'd like to believe that all these "theories" are political beliefs in the main. A liberal like Krugman who is so sceptical of Fama's efficient market theory and Lucas' rational expectations hypothesis is so willing to buy the prowess of the same inefficient market in predicting future Fed actions! It is convenient because the "market expectations" interpretation of the yield curve helps him shrug off critics who cavil at government spending and the huge debt burden by saying "Look...the bond market isn't particularly worried about the debt. Why bother!"
Similarly, the conservative Megan does not buy Theory B in the context of government bond market, but is quite likely to buy the not too dissimilar rational expectations theory which helps her make a case against expansionary government policy.
It's all so very politically convenient.