For some reason – maybe he is still annoyed with Scott Sumner – Paul Krugman decided to channel a post by Mike Konczal purporting to show that Market Monetarism has been refuted by the preliminary first quarter GDP numbers showing NGDP increasing at a 3.7% rate and real GDP increasing at a 2.5% rate in Q1. To Konczal and Krugman (hereinafter K&K) this shows that fiscal policy, not monetary policy, is what matters most for macroeconomic performance. Why is that? Because the Fed, since embarking on its latest splurge of bond purchasing last September, has failed to stimulate economic activity in the face of the increasingly contractionary stance of fiscal policy since them (the fiscal 2013 budget deficit recently being projected to be $775 billion, a mere 4.8% of GDP).
So can we get this straight? GDP is now rising at about the same rate it has been rising since the start of the “recovery” from the 2007-09 downturn. Since September monetary policy has become easier and fiscal policy tighter. And that proves what? Sorry, I still don’t get it. But then again, I was always a little slow on the uptake.
Marcus Nunes, the Economist, Scott Sumner, and David Beckworth all weigh in on the not very devastating K&K onslaught. (Also see this post by Evan Soltas written before the fact.) But let me try to cool things down a bit.
If we posit that we are still in something akin to a zero-lower-bound situation, there are perfectly respectable theoretical grounds on which to recommend both fiscal and monetary stimulus. It is true that monetary policy, in principle, could stimulate a recovery even without fiscal stimulus — and even in the face of fiscal contraction — but for monetary policy to be able to be that effective, it would have to operate through the expectations channel, raising price-level expectations sufficiently to induce private spending. However, for good or ill, monetary policy is not aiming at more than a marginal change in inflation expectations. In that kind of policy environment, the potential effect of monetary policy is sharply constrained. Hence, the monetary theoretical case for fiscal stimulus. This is classic Hawtreyan credit deadlock (see here and here).
If monetary policy can’t do all the work by itself, then the question is whether fiscal policy can help. In principle it could if the Fed is willing to monetize the added debt generated by the fiscal stimulus. But there’s the rub. If the Fed has to monetize the added debt created by the fiscal stimulus — which, for argument’s sake, let us assume is more stimulative than equivalent monetary expansion without the fiscal stimulus — what are we supposed to assume will happen to inflation and inflation expectations?
Here is the internal contradiction – the Sumner critique, if you will – implicit in the Keynesian fiscal-policy prescription. Can fiscal policy work without increasing the rate of inflation or inflation expectations? If monetary policy alone cannot work, because it cannot break through the inflation targeting regime that traps us at the 2 percent inflation ceiling, how is fiscal policy supposed to work its way around the 2% inflation ceiling, except by absolving monetary policy of the obligation to keep inflation at or below the ceiling? But if we can allow the ceiling to be pierced by fiscal policy, why can’t we allow it to be pierced by monetary policy?
Perhaps K&K can explain that one to us.
Uneasy Money 
Can fiscal policy work without increasing the rate of inflation or inflation expectations?
I can’t see why not. What do inflation expectations have to do with anything?
The underlying assumption of the multiplier is that economic actors are NOT in general operating on anything like an intertemporal budget constraint, but instead that current expenditures are closely linked to current incomes. This could be because liquidity constraints are pervasive, or because expectations are adaptive or backward looking, or because there are multiple equilibria, or for some other reason. The important thing is that in a world where current expenditure is tightly bound to current income, fiscal policy doesn’t have to change anyone’s expectations of anything in order to be effective.
The beginning of wisdom here, IMO, is to remember that intertemporal optimizing and budget constraints are analytic tools that may (or may not) be useful tools for analyzing certain questions. They are not true facts about the world.
LikeLike
Cool! Right through the ‘heart of the matter’.
LikeLike
David,
“Here is the internal contradiction – the Sumner critique, if you will – implicit in the Keynesian fiscal-policy prescription. Can fiscal policy work without increasing the rate of inflation or inflation expectations?”
First, it depends on the financing arrangement. FOMC enacted monetary policy involves the central bank buying government bonds in essence creating new money to buy the new debt. This assumes that the federal government must sell debt to fund deficits, which is itself a bad assumption.
Second, it depends on how the money is used. Obviously paying people to be unemployed ranks right up there with paying people to drop bombs and fire guns. The federal government (courtesy of the Humphrey Hawkins Act), has sworn off competing with private enterprise in the production of goods. If it backed off of that promise, then it could put downward pressure on prices.
LikeLike
Honest question: why does the Fed need to monetize the debt for fiscal stimulus to work?
LikeLike
Jon,
Honest answer: How would you measure the success of fiscal stimulus? What objective are you trying to obtain with fiscal stimulus?
LikeLike
David,
What do you mean by the process of monetization of debt here?
1) Central banks directly buying treasury bonds?
2) Direct credit from the central bank to the government agencies?
3) Printing cash and paying for the expenses in it?
If it’s none of the above, then I kind of don’t understand the exact meaning of the phrase.
LikeLike
“Can fiscal policy work without increasing the rate of inflation or inflation expectations?”
I think because fiscal policy can include direct investment by the govt (for example in infrastructure projects) it gives (in theory) a greater ability to stimulate RGDP within a given inflation target.
For example:
Monetary policy via QE will boost NGDP via a combination of inflation and RGDP growth and if you have an inflation target it will have limited scope.
Monetary policy via fiscal means (printing money to build bridges etc) will directly boost RGDP. As the money spent makes its way into the economy it will increase NGDP in the same way as for monetary policy. If inflation approaches target then traditional fiscal policy (tax increases etc) can be used to dampen demand and further infrastructure investment can be used to boost RGDP (and NGDP) if they start to fall below target..
Hence Keynesians would claim that fiscal policy can hit optimum RGDP with no inflation. However the price to pay for this is a larger state sector and a loss of market-driven outcomes.
LikeLike
Rob Rawlings,
You have hit the nail on the head. The problem becomes a political one – can an enterprise like the federal government limit itself to projects that boost real gross domestic product – instead of Congressman so and so voting for my bridge in exchange for his military base and deployment to Bum-F!ck-Egypt?
With a high political turnover, Congressional approval for even the most well intentioned infrastructure projects waxes and wanes. Keynesianism depends on a stable political backdrop because government expenditures other than interest on the debt are by their nature discretionary measures.
http://en.wikipedia.org/wiki/Party_divisions_of_United_States_Congresses
Democrats ruled the House of Representatives from 1955 thru 1995. Democrats ruled the Senate from 1955 thru 1981.
Since that time the House has swapped parties three times and the Senate has swapped parties once. Ever wonder why the Keynesian economists are focused on stimulus rather that permanently larger government sector?
LikeLike
Rob,
But usually a government investment doesn’t bloat a state sector that much, does it? I mean, if a government decides to build bridges, it could either create a big bureaucratic structure that will build everything and act as a firm in its own right (bloating the state sector and possibly stiffing the private initiative in bridge-building) or call a tender amongst the private contractors and simply pay the winner for the construction works (possibly growing the private sector). But since modern governments usually go the second rote, I guess the social investment is mostly safe.
LikeLike
Roman,
You are correct, but 1) the government is still making decisions about social utility that it may not be qualified to make and 2) it may make those decisions in a politically biased way. If one could avoid this by using purely monetary policy , and if the price is just a few % points of inflation for a year or 2 the trade off may be worth it.
LikeLike
And over here in the UK, 2% inflation has been more of a “floor” not a “ceiling” for the BoE, and yet people still take seriously the idea that the BoE is “impotent”.
LikeLike
And, very nice post!
LikeLike
robengland,
Despite the merits of your objections (jeez, do I feel the corruption and incompetence of Russian government), governments in a sense do exist to make decisions about social utility. There is something wrong in electing people to do stuff and simultaneously pressuring them to do nothing.
Besides, nobody is really qualified to make perfect decisions about anything. Ultimately, if a bridge is built, it was a decision of either a single person or a group of people (might be a corporation, some municipal community, might as well be a government). None of them are ideal, but still the bridges are built all around the world with mostly positive results. If anything, (federal) governments as nation-wide organizations are the most qualified to decide on social investment because they have the fullest picture, so to say.
On the question of whether the monetary policy may substitute the fiscal policy I am pessimistic. Setting a lower discount rate is impossible, doing more open-market operations is not very useful. The host of this blog suggested some time back that the most sensible way the Fed could intervene is by artificially lowering the value of dollar on the Forex market (doing currency interventions), but I think this is a dubious tool in a world where your trade partners may and will retaliate. And what remains? Changing the expectations of inflation and growth by public announcements? Doesn’t work at all, judging by the last few years.
Though there is also a possibility of debt jubilees (like those proposed by Steve Keen). The Fed could buy household debts, quickly grinding down the mountain of private debt that hampers the economy. But this probably would be very inflationary. What do you think of this option?
LikeLike
Monetary policy is out of control!
LikeLike
JW, You asked:
“What do inflation expectations have to do with anything?”
In the Keynesian model, Inflation expectations raise the marginal efficiency of capital and increase the cost of holding money, both of which tend to increase expenditures. An intertemporal budget constraint is not necessary to derive this effect. It was recognized by Keynes in the General Theory. Nevertheless, I take your point that by directly increasing expenditure, fiscal policy could increase employment without altering inflation expectations. That said, I am not so sure that you can argue that the increase in employment can take place without an increase in inflation, and if the central bank is targeting inflation will not an expansionary fiscal policy trigger a monetary tightening to prevent inflation from rising above its target?
Marcus, Thanks.
Frank, I am not necessarily saying that the government must finance its deficits that way, but that does seem to be the way it works out in practice.
Jon, If the government borrows without creating new money, then the money borrowed must reduce someone else’s spending unless increased interest rates induces some people to reduce their holdings of money. At the zero lower bound, interest rates don’t rise so money holdings are not reduced.
Roman, I mean option 1).
Rob, I agree that if fiscal policy can work without altering expected inflation, but that does not mean that it has no effect on inflation. Perhaps that gives fiscal policy a little more headroom than monetary policy in increasing aggregate demand without piercing the price level or inflation target, thereby postponing the monetary tightening triggered by violation of the price-level/inflation target, but my intuition is that the difference would not be all the large. I don’t see how fiscal policy can increase NGDP and reduce unemployment without inflationary effects. How do you increase the demand for labor (reduce unemployment) without increasing wages?
Roman, Your point is well taken. The size of the state sector is not unambiguously measurable. The government can build houses for people to live in or it can subsidize the rent or mortgage payments made by private individuals. In either case, the amount of government spending will rise by roughly the same amount, but the size of the state sector will be larger in the latter case than the former.
robengland, There are various layers of complexity here. It seems a perfectly reasonable decision for a liberal democratic society to make to ensure that certain kinds of services are available to its citizens, e.g., housing and education. Those decisions have to be made collectively. But there are lots of different ways of arranging for the delivery of those services. The government can provide the education through schools that the state operates or that are operated by state-supported entities like local school boards or it can give vouchers to parents that can be spent at schools of the parents own choosing. I used to think that it was obvious that vouchers were a superior strategy. I am no longer convinced that that is the case.
Britmouse, Because of indirect taxes, it is hard to measure UK inflation in a meaningful way. What has been happening to nominal wages in the UK? And thanks!
Roman, First, on currency depreciation. See my posts on competitive devaluation and currency manipulation. Those would be good a good thing; they would effectively raise the overall rate of inflation all over the world. Nominal exchange rate depreciation is not the same thing as real exchange rate depreciation. Real exchange rate depreciation requires a combination of foreign exchange rate intervention and sterilization via tightening monetary policy. No one is suggesting such a policy. As for the Fed buying household debt, if the Fed can increase the price level by buying government debt, the burden of household debt will automatically be reduced, so there is really no point in doing that until the Fed has bought up all the government debt.
Tas, In the US, monetary policy is legally under the control of the FOMC, whose membership is determined under legislation enacted by Congress. So in a legal sense, your assertion is not true. If you mean that the FOMC is not doing its job properly, I agree. But I think that we disagree about what the FOMC should be doing. That is a question of policy, not anarchy.
LikeLike
David,
“How do you increase the demand for labor (reduce unemployment) without increasing wages?”
You change the compensation from highly liquid assets (currency) to less liquid assets. Private companies do it all the time (deferred compensation) in the form of options (usually equity options), retirement benefits, and the like.
In that way you are affecting the liquidity preference of labor. Which is not to say that labor of its own accord cannot decide how much to save and how much to spend of available income.
LikeLike
David,
You write: “In principle [fiscal policy] could [help] if the Fed is willing to monetize the added debt generated by the fiscal stimulus.”
But why is monetization required if by monetization you understand direct funding of Treasury by the Fed? In the current institutional setups around the world, this mode of central banks operations is usually explicitly forbidden – the system is designed in such a way that Treasuries have to sell their bonds to the private banks first and a central bank will buy them from the banks later. Your position, if I understand you correctly, is more extreme than even that of the MMT movement: they think that debt monetization would be more straightforward but essentially equal way of operating the Fed, you suppose that it is required for the activist fiscal policy. But why? Marc Lavoie [2011] shows that there is no major difference to the outcome whether the government finances itself by selling bonds to the private banks first or directly to the central bank.
I am afraid that my grasp of the international trade mechanics is shaky indeed, so I’ll have to re-read your previous posts on this matter. But even if currency manipulation could be inflationary, inflation by itself is only a very indirect remedy to the problem of people lacking income or being stiffed by debt (that in extreme cases is even indexed by inflation). Sure, it could make an economic situation better, but what a timeframe for that would look like? Debt jubilees, on the other hand, are rather instantaneous instruments (and could provide even more inflation).
LikeLike
Roman P,
I can see why the government borrowing from an individual bank (commercial or central) is inflationary. I do not see why the government borrowing from the banking system is inflationary. As a partial equilibrium fact, a sale of a government bond to a bank increases the money supply. I do not see why that leads to a general equilibrium result of an increase in the money supply.
(I’m not being rhetorical: I’m sincerely trying to work out the arguments on both sides here, having previously strongly held the view that such transactions between the government and the financial system are inflationary even once all agents have responded to the transaction.)
LikeLike
Frank, Can you explain what you mean in very concrete terms? Otherwise, I don’t think I can grasp what you are trying to say.
Roman, You misunderstood me. By directly buying Treasury bonds, I did not mean that the Fed has to buy Treasury bonds from the Treasury, just that they buy an equivalent amount of bonds to the amount of the borrowing that the Treasury undertakes. I am not sure what your question is concerning currency manipulation and inflation. Perhaps you can ask it again so I can follow it.
W. Peden, Your first two sentences are completely incomprehensible to me. Why is buying from the entire banking system less inflationary than from a single bank?
LikeLike
David Glasner,
I should have expressed myself much more clearly: an individual bank may create a deposit, but in doing so it raises the banking system’s demand for base money (since a deposit is a claim on base money) and if this demand is not accommodated by the central bank with new base money, then the increase in the money supply must be reversed.
So the individual transaction between the government and the bank in the bond purchase is inflationary, but (in itself) the transaction has no inflationary effects in aggregate whatsoever. So what is true of the individual case is not true of the aggregate case.
(None of this assumes that the demand of banks for base money relative to their deposits is constant or that central banks always target quantities of base money.)
LikeLike
David, first of all, I love reading your posts as you’ve an innate ability to articulate complex economic concepts in a simple language, which helps people like me, who have no economics background, to understand the myriad issues that currently mire the global economy.
From my limited understanding, the point that Krugman is making is aptly summarised in his post at the end, i.e., “monetary policy can, for the most part, gain traction under current circumstances only by changing expectations about future actions (and changing them a lot). Meanwhile, fiscal policy has a direct, current effect on the economy”. The answer to your point about the internal contradiction that as to how fiscal policy can work without increasing the rate of inflation or inflation expectations, is, that it cannot. But then the increase in inflation is a consequence of fiscal activism and not an objective in itself. On the other hand, for monetary policy to be effective, as you said, “ it would have to operate through the expectations channel, raising price-level expectations sufficiently to induce private spending.” Therefore, piercing of the 2% inflation ceiling becomes an objective of monetary policy and that in turn impacts inflation expectations and induce private spending. My understanding is that it takes time for monetary policy to be effective on account of lags in transmission. Therefore, while we continue to pursue a loose monetary policy, it is also important to understand that it may not completely offset the fiscal contraction. Therefore, as Krugman says, when the Government is not in competition with the private sector, Government purchases don’t use resources that would otherwise be producing private goods and Government borrowing doesn’t crowd out private borrowing, why not use both fiscal and monetary policy to try to help all those millions who are struggling to make a living.
LikeLike
W. Peden,
I am not sure that I follow your question enough to provide you a satisfactory answer, sorry. But your distinction between the partial equilibrium and general equilibrium facts is certainly misleading – better to look at the universal facts of accounting.
David,
Thanks, now I understand your position.
LikeLike
Roman P,
That sounds like a terrible idea, if it involves giving up the distinction between partial equilibrium and general equilibrium.
LikeLike
David, UK nominal wages have grown somewhere in the 0-2% yoy range for most of the last three years, depending on what measure you use.
Ironically this is one of things the BoE see this as a “success”, after the 25% Sterling devaluation they had to make sure we didn’t rerun the 1970s.
LikeLike
David,
“Frank, Can you explain what you mean in very concrete terms? Otherwise, I don’t think I can grasp what you are trying to say.”
Certainly. Here is what you said:
“How do you increase the demand for labor (reduce unemployment) without increasing wages?”
Here is what I answered:
“You change the compensation from highly liquid assets (currency) to less liquid assets that offer a rate of return.”
Wages are compensation in a liquid asset (currency) that is the government’s liability. Other forms of compensation include options (equity or bond), retirement benefits, healthcare, etc.
In addition wages are funded out of a company’s operating cash flow. And so you have two competing interests – the company who wants to use its limited cash flow to hire more workers and workers who want more in compensation as labor demand increases.
A company in this case can reward its employees in an asset that makes a tradeoff between liquidity and return on investment. That asset can be a liability of the company (options for instance) or it can be a liability of the government.
LikeLike