One of the big insights of macroeconomics in the last 40 years or so is that, where possible, demand management is best left to monetary rather than fiscal policy. Incredibly, this reasoning successfully permeated the thinking of policy makers in this country and elsewhere.
The only time you will find any mainstream academic economist supporting a fiscal stimulus is at times like the present, when short-term interest rates are close to zero. In this setting, some New Keynesian models can be used to support fiscal stimulus, and this analysis has been widely used by opponents of the government’s actions to reduce the deficit.
It is important to realise how reliant this New Keynesian case for fiscal stimulus is on monetary policy being powerless to manage the economy: once we have returned to a more conventional monetary policy regime, the case for fiscal stimulus will be severely weakened. Furthermore, with inflation above target, the end of QE in sight and movements in bond markets suggesting that traders expect tighter monetary policy, it seems that this return to normality could occur sooner rather than later and probably before the time of the next election.
As a result, we should see greater consensus from the left of British politics on the necessity of fiscal adjustment. Unless of course, New Keynesian models only guide their thinking when they promote higher government expenditure?
The New Keynesian Case for Fiscal Stimulus
We disagree with the New Keynesian case for fiscal stimulus for a variety of reasons, which include:
The point we want to make here is that, even when we ignore all of these concerns, the theoretical case for fiscal stimulus relies to an extraordinary extent on interest rates being stuck at zero.
To outline the New Keynesian argument let’s look at one of the key pieces of academic research used by proponents of fiscal stimulus: “When is the government expenditure multiplier large” by Professors Christiano, Eichenbaum and Rebalo of Northwestern University.
This is a typical New Keynesian model: it consists of rational consumers who choose how much to save or consume, rational firms that occasionally change their prices to maximise their profits and a central bank that sets nominal interest rates to stabilise the economy. Crucially, nominal interest rates are, as in reality, constrained to never fall below zero.
Fiscal Multipliers in Normal Times
In normal times (when nominal interest rates are positive) fiscal multipliers are small, with a typical value being in the region of one. A value of less than one, but more than zero implies that increasing government expenditure by one pound increases output by less than one pound. In other words it acts to reduce private sector output and this “crowding out” effect offsets the effect of fiscal stimulus.
Fiscal Multipliers at The Zero Lower Bound
The authors argue that all this changes once interest rates are at the zero lower bound: fiscal multipliers can easily be two or more, some specifications deliver multipliers exceeding twenty. This implies that fiscal stimulus could easily pay for itself!
Why the Distinction?
To understand where this distinction comes from it is first necessary to understand what New Keynesians think of as a recession. In this, and most other models making the case for fiscal stimulus, large recessions are seen as resulting from “shocks” (like the recent financial crisis), which cause a sudden increase in consumers’ desire to save.
In normal times, these shocks can be alleviated by the central bank: by cutting nominal interest rates it reduces the return to saving, thereby stimulating consumer spending and staving off recession. In this setting, fiscal stimulus is unnecessary. Intuitively, the Central Bank knows the amount of stimulus required and so, if government gets involved with fiscal stimulus, the Central Bank will offset it by making less use of monetary stimulus.
In contrast, when the initial shock is so large that the central bank would like to choose a negative interest rate, but cannot do so because of the zero lower bound, fiscal stimulus does the central bank a favour: it delivers stimulus that the central bank is itself powerless to provide. As a result, there is no offsetting increase in interest rates and fiscal multipliers can be large.
What Does this Mean for Policy?
The above argument represents the primary theoretical case for fiscal stimulus. We at the CPS believe it to be flawed for the reasons I outlined at the beginning. But, when a politician calls for fiscal stimulus, this is the logic they are appealing to (whether they realise it or not).
As long as monetary policy makers are happy with the current level of stimulus they are providing (and indeed markets seem to believe that tighter policy is on its way) the above logic does not apply: a fiscal stimulus will be offset by a faster tightening of monetary policy. This implies that fiscal multipliers are likely to be small, probably less than one, and nowhere near large enough to make stimulus self-financing.
Furthermore, if nominal interest rates have risen by the time of the next election, then there really should be no question as to whether action to cut the deficit is necessary: even New Keynesian’s believe that stabilization policy is best left to monetary policy during normal times and that fiscal adjustment is necessary in the medium term. We hope that those on the left of the British policy debate are as eager to listen to this aspect of the New Keynesian analysis as they were to the part that supported increased government expenditure!