Channel Surfing with the Fed

To understand what’s driving US monetary policy you must be a student of ‘Fed Speak.’ Today’s ‘Fed Speak’ incorporates ‘channels,’ ‘stories,’ ‘core’ this and ‘core’ that, with expressions such as ‘accommodation,’ ‘neutral rate,’ and the like piled on.  So fasten your seatbelt and enjoy the ride!

The dominant instrument of monetary policy continues to be the manipulation of the fed funds rate, with the presumption that ‘inflation’ is a function of ‘slack,’ a.k.a. the ‘output gap,’ which is the difference between potential GDP (gross domestic product) and actual GDP, and that interest rates influence the level of economic activity through the ‘demand channel’ and the ‘expectations channel.’  This is a big change from the ‘70s and ‘80s when the Fed attempted to control the ‘money supply’ to influence the economy.

As Chairman Greenspan recently proclaimed, ‘inflation is a monetary phenomenon.’  In late 2004, this means that ‘core CPI’ (consumer price index) is a function of the ‘real rate of interest,’ which is the nominal rate of interest minus ‘core CPI.’  So, for example, with the fed funds rate at 1.5%, and core CPI at 2.0%, the real rate of interest is negative .5%.  The Fed has made it clear that a negative ‘real rate’ is ‘accommodative,’ and that, while this conclusion implies that there is a ‘neutral rate,’ the actual ‘neutral rate’ is not known.  ‘Neutral’ is defined as a rate that is neither ‘accommodative’ nor ‘contractionary.’  In subsequent elucidation, Fed members have indicated today’s ‘neutral’ fed funds rate may be anywhere from 3.5% to 5.5%, but they always add that they aren’t sure exactly what it is.

 Next let’s take a look at ‘inflation.’  When it comes to the CPI, the broadly accepted measure of ‘inflation,’ the Fed is quick to recognize a ‘relative value story’ such as a ‘supply shock,’ and exclude it from ‘core inflation.’  For example, suppose a supply shortage of crude oil causes the price of crude oil to increase.  This is a ‘supply shock’ and not, per se, ‘inflation.’  It is simply a case of ‘markets functioning’ to allocate scarce resources by price, which also allows the price to seek a level that brings out new supply.  (The Fed is ever mindful of whether markets are ‘functioning’ or not) The higher price of oil may also put downward pressure on other prices, as the extra spending power used to purchase oil reallocates demand away from other goods and services.  So the Fed designs ‘core CPI’ and ‘core deflators’ to detect ‘core inflation’ which is a ‘monetary phenomenon,’ and exclude ‘relative value stories’ and ‘supply shocks.’

 ‘Inflation,’ the ‘monetary phenomenon,’ works via the ‘demand channel.’  (It does not work through the ‘supply channel,’ which is a ‘relative value story.’)  If the ‘real rate’ is ‘accommodative’ (too low), ‘core inflation’ will face upward pressure, as the profit motive encourages agents to borrow and spend.  The exact transactions agents will undertake to profit by doing this, however, can’t be specified, as one can’t directly purchase the elements of ‘core CPI.’

 Inflation can also be propagated via the ‘expectations channel.’  The theory is that the presence of a ‘negative real rate’ raises ‘inflationary expectations.’  This theory is evidenced by increased buying of goods and services as price increases are anticipated, and increased demands for higher wages due to increased ‘inflation expectations.’  And it is wages- with emphasis on unit labor costs- that constitute the key element of ‘core inflation.’  (By the way, wages are never a ‘relative value story’)  This relationship introduces the ‘productivity story’ which tells the tale of obtaining more and more output from the same number of employees.  Many in the Fed will attribute the apparent shift in the ‘speed limit,’ which is the presumed maximum GDP growth that can be safely tolerated without inducing ‘core inflation,’ to the ‘productivity story.’  Economists  once thought that 3% might be the ‘speed limit’ but the last few years they have raised that view to at least a 4% growth rate of GDP, provided the ‘productivity story’ continues.  Of note at the last Congressional hearings were the likes of Senator Kennedy and Congressman Rangle commending Fed Chairman Greenspan for doing a good job, after the Fed Chairman explained that ‘core inflation’ was under control due to low unit labor costs and a lack of wage pressures.  Who would have thought the day would come when the American left would congratulate a Fed Chairman for effectively suppressing wages!!!???

So today, the Fed is in the process of ‘removing accommodation’ as the negative ‘real rate’ implemented to address concerns of a widening ‘output gap’ and excess ‘slack’ have now increased the ‘upside risk’ to the economy, as ‘accommodative’ ‘real interest rates’ are causing both an increase in demand that could close the ‘output gap’ too quickly (the Fed calls this ‘the first derivative of slack’), resulting in inflationary pressures as one moves along the ‘Philips Curve’ and approaches the ‘NAIRU’ and an increase in ‘inflationary expectations.’

The ‘NAIRU???’  Yes, there’s another construct vital to understanding the ‘demand channel’ of monetary policy.  The NAIRU is short for the ‘Non Accelerating Inflation Rate of Unemployment.’  What this definition means is that there is some level of ‘unemployment’ above which there is downward pressure on wages and below which there is upward pressure on wages.  As with the ‘neutral’ rate of interest, there is an implied level of the NAIRU that is neither inflationary nor deflationary, that can be estimated and used as a policy input (Not long ago Chairman Greenspan estimated it at 4%).  Also, if one drew a graph with unemployment on one axis and ‘core inflation’ on the other, it would take the shape of a curve called the ‘Phillips Curve’ with higher levels of unemployment resulting in lower ‘core inflation’ and vice versa.

 The Fed is currently in the process of ‘removing accommodation,’ as described above.  But, it explains, due to the relatively weak economy, modest GDP growth, sufficient economic ‘slack’ (high unemployment and low capacity utilization), and relatively low levels of ‘core inflation,’ and because energy is a ‘relative value story,’ it is ‘removing accommodation’ only ‘gradually.’ This ‘gradualism’ has meant increasing the fed funds rate by ¼% increments at the last two Fed meetings.  Presumably this rate of increase will continue until they reach the ‘neutral rate’ which currently is unknown.  The Fed has used weakness in the economy to support its decision to move ‘gradually,’ rather than in increments greater than ¼%.  But in no way have they shown any inclination to not ‘remove accommodation,’ as they are ever mindful of the ‘expectations channel,’ which means they must act and talk tough to keep expectations of ‘future inflation’ at bay.  And it also helps to factor in that the Fed tends to be willing to ‘take chances’ with a higher GDP growth rate when inflation is falling, and be more tolerant of weaker GDP growth when inflation is rising, as is now the case.

Monetary policy seems simple.  After a period of decline, ‘core cpi’ was finally responding to ‘accommodative negative real rates’ and moving up, and the ‘output gap’ was beginning to reverse course as well, as indicated by higher job growth and what looked like a bottoming of capacity utilization and a reduction in productivity growth.  The jobless recovery, a ‘productivity story,’ showed signs of improvement.  ‘Slack’ in the economy was abating.

But, alas, there’s more.  The Fed has a much more sophisticated level of understanding layered on the above premises.  We must now add the ‘long term neutrality of money.’  This means monetary policy, in the long run, does not alter output and employment.  Therefore interest rate changes, through the ‘demand channel,’ only alter ‘core inflation’ by moving spending forward or backward to or from future periods.

This view is consistent with fact that for every $ borrowed there is a $ saved.  This identity mean’s that while higher interest rates discourage borrowing, they also encourage spending, as savers earn exactly the extra interest borrowers must pay.  Consequently, at the ‘macro level,’ what matters is only the difference in the ‘propensity to spend’ of borrowers vs. savers.  This is not to say changes in interest rates will not have large ‘micro’ results.  Indeed, the Fed closely watches these micro effects as they may have large ‘multipliers’ and alter the ‘macro economy’ via disproportionate ‘propensities to spend’ of particular sectors.

And here, surprise!, we introduce still another channel- the ‘fiscal channel.’  In fact, recent Fed studies (May, 2003) showed reasonably strong correlations between fiscal balance and GDP.  Tax cuts and spending increases necessarily channel exactly that much additional net income and holdings of financial assets to the non-government sectors.  And Fed analysts are quick to point out that the relatively large deficits of the third quarter of 2003’s retroactive tax cuts may be what was responsible for subsequent growth via this ‘fiscal channel,’ and not the level of ‘real interest rates.’  Therefore, there they are cautiously optimistic on the economy, as it is possible the ‘accommodative’ interest rates may have already done their part through the ‘demand channel’ to move spending forward, while ‘demand stimulus’ via the ‘fiscal channel’ is waning as well.  This would mean an economy with deficient levels of aggregate demand, and no immediate means of the Fed increasing demand sufficiently to reduce the ‘output gap.’

Dare we say ‘Japan, ten years ago?’  Note that the Fed has been quick to say it learned the lesson of Japan.  And to demonstrate this, the Fed lowered rates more rapidly than Japan had done 10 years earlier in reaction to a similar equity market bust.  However, so far in 2004, the Fed has raised rates from 1.00% to 1.5%.  And this happens to be exactly what Japan did in 1994, when they raised their interbank rate from 2% to 2.5%, only to subsequently realize that miscalculation and reverse course.  Nearly 10 years of near 0% rates followed, in a decade long fight against deflation.

Now consider some recent work relevant to this discussion that has been done by Professor Stephanie Bell at UMKC who advanced evidence that, with high enough government deficits, raising rates can actually be expansionary, as increased interest payments by government working through the ‘fiscal channel’ overcome any macro influences of the higher interest rates through the ‘demand channel.’  The reverse is true as well.  Lower interest rates reduce government spending on interest payments, and, again, the ‘fiscal channel’ can be more powerful than the ‘demand channel’ if the government debt is large enough.  And Philip Arestis has also found very low correlations at best when try to determine the effect of interest rate changes in the Eurozone.

The concept of a NAIRU is also under heavy attack.  For starters, any student of game theory will tell you that the ‘labor market’ can’t be a ‘fair game’ as people need to work to survive and business needs to hire only when it determines doing so is sufficiently profitable.  Additionally, all the empirical research shows mixed results at best.  This makes it difficult for the Fed to know how large the ‘output gap’ actually is, and what the ‘speed limit’ for the economy is.

And, as the last complexity for today, there is a very strong ‘supply side story’ that explains how higher interest rates raise the cost of investment, and therefore the CPI must go higher before new supply comes into the market.

So what should a voting Fed member do?  As he surfs though the channels, a very confusing series of pictures flashes on his screen.  And it’s not even clear whether raising rates increases or decreases ‘inflation.’

Maybe Japan has it right?  Set the funds rate at 0 and leave it there.  There is no evidence that this causes ‘inflation’ or any other ‘bad’ thing, and ‘savings’ of financial assets have actually gone up, not down, through the ‘fiscal channel.’  Low rates cut government spending on interest, mandating offsetting spending increases or tax cuts to avoid contraction via the ‘fiscal channel.’  Purchasing power is shifted from ‘savers’ to working people.  Lower long term rates encourage investment, as asset markets make a one time adjustment to the lower level of interest rates and ‘markets function’ to allocate resources.

Think about it!

Philip Arestis
Cambridge Centre for Economic and Public Policy
University of Cambridge
Levy Economics Institute.

Warren Mosler
Associate Fellow
Cambridge Centre for Economic and Public Policy
University of Cambridge.

Address for both authors:
Cambridge Centre for Economic and Public Policy
Department of Land Economy
University of Cambridge, 19 Silver Street
Cambridge CB3 9EP

Posted September 2, 2004