With Bernanke's $40 billion a month promise...and the famous Woodford paper on which much has been written about, I was left wondering when i glanced at an article by Blackrock's Peter Fisher which is based on the threat of a liquidity trap, you know, the one that Keynes conceptualized. Fisher's point can be found in the first line itself.
The. Federal. Reserve. Should. Stop. Trying. To. Engineer. Lower. Long-Term. Interest. Rates
The reasoning is straightforward, i.e: that the lower long-term rates go, the more lending would be discouraged. That basically, there "would be no reward for those willing to give up current consumption or liquidity" - cash becomes attractive and rates this low would not positively impact credit creation but rather, have a perverse effect on lending and investment.
But the US economy IS in a liquidity trap. Conventionally, a liquidity trap would occur when increased money supply would fail to lower interest rates. On the ground, people would hoard cash based on expectations of deflation, depressed demand, war etc. Mathematically, it refers to a state in which the nominal interest rate is close or equal to zero effectively "zero-binding" conventional monetary policy. As it goes, a deflationary environment can arise thus creating a vicious cycle of output stagnation and further deflationary expectations. Furthermore, in a leveraged sector, the real value of debt rises leading to more balance sheet pain.
Using a Taylor Rule, like most before me I'll use the Mankiw version with an indicator at -8.5, i could run the numbers using BLS data and here's my chart with the Mankiw rule and the actual fed funds rate:
2009 is where the disparity starts. It's also interesting to note that sometime back in January, the economist ran an article in Free Exchange which claimed that the Mankiw rate would be positive in no time but that this steep is unlikely to continue. Here's a graph of the same data only from 2009 onwards (the effective fed funds rate is the right scale):
The problem is that Krugman re-estimates the coefficients (it's still a linear combination of
the unemployment/inflation differential), but his data is fresher and the results makes these look significantly overestimated. Hmmm. Four more years of zero-bound flirtation if you extrapolate using conservative CBO projections.
It's easy to hear the chorus singing the "unintended consequences" of the easing. By driving down long-term rates and easing mortgage pressure among others, the Fed, all said and done is making clever use of its balance sheet. In Bernanke's defense, it's hard to predict investment behavior in the absence of any easing. The zero bound constrains options because quite simply, when banks lend money to one another in the fed funds market, lenders simply don't pay for the "lending privilege". Hence, the fed is definitely not going to raise rates anytime soon. Mathematically, you would need (ballpark estimates) a percentage point drop in the unemployment rate coupled with near 1% inflation and even with a 2% inflation rate, there would ened to be a drop to 8% in the unemployment rate. Not happening. Zero-rate environment where conventional monetary policy will have minimal stimulatory effect.
Also, from an IS-LM view, expansionary fiscal policy would shift the conventional IS to the right thus putting upward pressure on interest rates - the 'crowding out' effect. But in a liquidity trap, the LM curve is flat and only curves towards the end of the output gap (full employment cuts through the LM curve but the economy isn't there). So a shift in IS has its effect on output but no push on interest rates.
The relationship between long-term yields and short-term interest rates brings out critics in full force. Fisher argues that lower yields forces investors to notice the smaller opportunity cost of holding cash. Simply put, if the Fed continues to hoard securities, investors might be forced to replace such assets with comparable risk leading to a further lowering of rates, easier lending conditions etc. What Fisher says is that portfolio shifts are far more ambiguous, i.e: it could spark a chase for yield and a path of increased (and perhaps abrupt) expected reversal in trend.
By the way, this is what the long-term US government bond yield and the Fed Funds rate differential looks like historically. I vaguely attempted to shade in recessionary periods but they're not perfectly accurate:
A more exact approximation would be helpful but at a glance, it seems better to focus on the pre-recession periods which seem to be associated with a kind of convergence or a downward trend. A convergence would imply either a drop in long-term treasuries or a rise in the fed funds rate (tighter monetary policy). What's interesting is that from 2009 onwards the significant movement in the differential can be broadly attributed to what's happening with the long term treasuries, the effects of easing.
It seems, from an aesthetic perspective, that most times there's a convergence to zero, a grey area follows but in a depressed economy with near-zero rates, the differential is simply the long-term treasury yield assuming no increases in the fed funds rate for the next couple of years at least.
The. Federal. Reserve. Should. Stop. Trying. To. Engineer. Lower. Long-Term. Interest. Rates
The reasoning is straightforward, i.e: that the lower long-term rates go, the more lending would be discouraged. That basically, there "would be no reward for those willing to give up current consumption or liquidity" - cash becomes attractive and rates this low would not positively impact credit creation but rather, have a perverse effect on lending and investment.
But the US economy IS in a liquidity trap. Conventionally, a liquidity trap would occur when increased money supply would fail to lower interest rates. On the ground, people would hoard cash based on expectations of deflation, depressed demand, war etc. Mathematically, it refers to a state in which the nominal interest rate is close or equal to zero effectively "zero-binding" conventional monetary policy. As it goes, a deflationary environment can arise thus creating a vicious cycle of output stagnation and further deflationary expectations. Furthermore, in a leveraged sector, the real value of debt rises leading to more balance sheet pain.
Using a Taylor Rule, like most before me I'll use the Mankiw version with an indicator at -8.5, i could run the numbers using BLS data and here's my chart with the Mankiw rule and the actual fed funds rate:
2009 is where the disparity starts. It's also interesting to note that sometime back in January, the economist ran an article in Free Exchange which claimed that the Mankiw rate would be positive in no time but that this steep is unlikely to continue. Here's a graph of the same data only from 2009 onwards (the effective fed funds rate is the right scale):
The problem is that Krugman re-estimates the coefficients (it's still a linear combination of
the unemployment/inflation differential), but his data is fresher and the results makes these look significantly overestimated. Hmmm. Four more years of zero-bound flirtation if you extrapolate using conservative CBO projections.
It's easy to hear the chorus singing the "unintended consequences" of the easing. By driving down long-term rates and easing mortgage pressure among others, the Fed, all said and done is making clever use of its balance sheet. In Bernanke's defense, it's hard to predict investment behavior in the absence of any easing. The zero bound constrains options because quite simply, when banks lend money to one another in the fed funds market, lenders simply don't pay for the "lending privilege". Hence, the fed is definitely not going to raise rates anytime soon. Mathematically, you would need (ballpark estimates) a percentage point drop in the unemployment rate coupled with near 1% inflation and even with a 2% inflation rate, there would ened to be a drop to 8% in the unemployment rate. Not happening. Zero-rate environment where conventional monetary policy will have minimal stimulatory effect.
Also, from an IS-LM view, expansionary fiscal policy would shift the conventional IS to the right thus putting upward pressure on interest rates - the 'crowding out' effect. But in a liquidity trap, the LM curve is flat and only curves towards the end of the output gap (full employment cuts through the LM curve but the economy isn't there). So a shift in IS has its effect on output but no push on interest rates.
The relationship between long-term yields and short-term interest rates brings out critics in full force. Fisher argues that lower yields forces investors to notice the smaller opportunity cost of holding cash. Simply put, if the Fed continues to hoard securities, investors might be forced to replace such assets with comparable risk leading to a further lowering of rates, easier lending conditions etc. What Fisher says is that portfolio shifts are far more ambiguous, i.e: it could spark a chase for yield and a path of increased (and perhaps abrupt) expected reversal in trend.
By the way, this is what the long-term US government bond yield and the Fed Funds rate differential looks like historically. I vaguely attempted to shade in recessionary periods but they're not perfectly accurate:
A more exact approximation would be helpful but at a glance, it seems better to focus on the pre-recession periods which seem to be associated with a kind of convergence or a downward trend. A convergence would imply either a drop in long-term treasuries or a rise in the fed funds rate (tighter monetary policy). What's interesting is that from 2009 onwards the significant movement in the differential can be broadly attributed to what's happening with the long term treasuries, the effects of easing.
It seems, from an aesthetic perspective, that most times there's a convergence to zero, a grey area follows but in a depressed economy with near-zero rates, the differential is simply the long-term treasury yield assuming no increases in the fed funds rate for the next couple of years at least.
No comments:
Post a Comment