Wednesday, 17 June 2015

Modern monetary theory and Simon Wren-Lewis.

Summary:      In as far as government incurs debt and runs deficits to deal with lack of aggregate demand, we should never have debt or deficit reduction targets. The debt and deficit should simply be whatever maximises employment without the inflation target being exceeded by too much. Moreover, deficit adjustments (and the debt adjustments that result from that) should ideally be the ONLY tool used to adjust demand.  That is, interest rate adjustments should only be used in an emergency.


Simon Wren-Lewis is an Oxford economics prof and I’ve never been totally happy with his analysis of national debts and deficits, thought I fully back his criticisms of what he calls “macromedia. That’s the simple minded equating of government debt and deficits with the debt and deficits of microeconomic entities like a household or individual person.

That conflation is a classic example of one of the most popular mistakes in economics: applying microeconomic thinking or laws at the macroeconomic level. And the majority of politicians and perhaps roughly half of economics commentators working for newspapers make that mistake.

With a view to trying to show how modern monetary theory (MMT) is a bit better than SW-L theory here, I’ll start with the MMT take on national debts and deficits – at least what I think the MMT take is.

There is of course no central MMT authority to turn to get the “official” MMT view here, any more than there is one central authority for the official Keynsian, monetarist, or Austrian view on anything. But I suspect most MMTers will agree with my suggestions as to what the MMT take is.

Be warned however: setting out MMT thinking takes about a thousand words below before we turn to showing how that is a bit better than SW-L thinking.

Also, and to repeat, I’m not saying there is a FUNDAMENTAL difference between MMT and SW-L. I’m dealing with minor differences. But it’s of benefit to get everything 100% right in this area, as thousands of jobs depend on getting things 100% right.

Another important initial point is that the discussion below refers to a country that issues its own currency. I.e. the discussion does not apply to a country in the Eurozone, though it does apply to the EZ as a whole.

Private sector net financial assets.

MMTers often use the phrase “private sector net financial assets” (PSNFA), and PSNFA is made up of two elements: base money and national debt. The two latter have in common the fact that they are ASSETS as viewed by the private sector entities that hold them.

Moreover, those two parts of PSNFA are very little different in nature, particularly at low interest rates. That is, $X of national debt is simply a promise by government to pay the holder of the debt $X on a particular date. And assuming that’s in the near future, then there isn't much difference between $X and a promise by government to pay you $X in the near future.

Indeed, as Martin Wolf (chief economics correspondent at the Financial Times) put it, “Central-bank money can also be thought of as non-interest-bearing, irredeemable government debt. But 10-year Japanese Government Bonds yield less than 0.5 per cent. So the difference between the two forms of government “debt” is tiny…”.

Keynes’s paradox of thrift.

Next, MMTers fully endorse Keynes’s point about what he called the “paradox of thrift”. That’s the fact that if private sector entities (households and employers) save money, they are ipso facto not spending that money. And any cut in spending raises unemployment.

MMTers claim that it is therefore important for the private sector to have the PSNFA it wants. Put another way, it is important to meet what MMTers often call the private sector’s “savings desires”.

The size of PSNFA is related to interest rates.

The next important point is that the amount of PSNFA that the private sector will want to hold is related to the rate of interest offered on that debt (a very obvious point, but one which is often overlooked). I.e. if someone is offered a return of 10% for abstaining from consumption and lending $Y to government (or any other entity), they are more likely to do that than if they are offered 1%.

What’s the point of government debt?

A possible reason for government debt is to fund infrastructure. That idea is actually very debatable. That is, there are good arguments for funding public investments out of tax rather than borrowing. (There’s a paper by Kirsten Kellerman in the European Journal of Political Economy arguing the latter point).

But let’s ignore government debt used to fund infrastructure, to keep things simple. Or if you like, assume that if indeed government DOES fund infrastructure via debt, that is kept separate from other government debt. The interest on that “infrastructure debt” might be same as would be charged by those funding a private contractor for constructing a road and charging a toll to those using the road.

So…to re-phrase the above question: what’s the point of government running up debt simply because it has failed to collect enough tax to cover its current spending (as opposed to capital spending)? Indeed most people are not too bothered about government debt where it is incurred to fund productive capital investments: it’s the debt incurred simply because government has failed to collect enough tax to fund CURRENT spending that is the major concern.

Well the answer to the above question is surely that there’s absolutely no point in failing to collect enough tax to cover current spending. Thus the only merit in government debt is to supply the private sector with PSNFA.

Why pay interest on PSNFA?

But why pay any interest on national debt when the private sector positively wants it as a form of saving? That is, why not just have PSNFA made up of base money rather than base money PLUS government debt on which government pays interest. Milton Friedman and Warren Mosler argue for that zero debt arrangement. That is, government debt might as well be abolished, which would mean the only liability (aka PSNFA) issued by the state would be base money.

Alternatively, it could be argued that it’s desirable to have PSNFA partially in the form of debt which pays some interest. Reason is that base money is potentially volatile stuff (which is presumably why it is sometimes called “high powered money”): holders of that form of money can always go into a fit of irrational exuberance and try to spend away large chunks of their base money. In contrast, it is more difficult to spend away debt: try doing your weekly shopping at the supermarket with $1k of Treasuries (or in the case of the UK, £1k of Gilts).

However, answering the latter question is not actually crucial to the basic question addressed here, so let’s just assume that SOME PSNFA is in the form of debt and that it’s very nearly the same thing as base money because as in Japan, the rate of interest is near zero. (There’s certainly little point in paying anything MUCH above zero, as explained above).

The optimum amount of debt.

We can now answer the big question: what’s the optimum amount of debt (aka PSNFA)? Well the answer is simple: whatever amount brings full employment when interest on the debt is near zero.

Note incidentally that that is NOT TO suggest, as extreme monetarists do, that the quantity of money should be the only thing varied when it’s desirable to adjust aggregate demand. Reason is that THE PROCESS of adjusting PSNFA also has an effect on demand: what might be called a “fiscal” effect. For example if there is insufficient demand, the state can simply create new money and spend it (and/or cut taxes) and the fact of that spending (e.g. on health and education) will create nursing and teaching jobs. Thus there is a fiscal and monetarist effect there.

Thus a better answer to the question as to what the optimum amount of debt is thus. If the economy is not at capacity (and assuming no inflation and no growth) we should create new base money and spend it, and/or cut taxes, and continue doing so till the size of the debt / PSNFA is such that the private sector spends enough to give us full employment.

In the real world of course we have both inflation and growth. The former tends to erode the real value of PSNFA which means that assuming (to keep things simple) the private sector’s desired PSNFA/GDP ratio is constant, fresh base money will have to be CONSTANTLY created to keep that ratio constant. And much the same applies to growth: that is to keep the latter ratio constant, there will need to be a constant supply of new base money.

Indeed that constant supply of new base money is what we have actually experienced over the last century or two.

SW-L thinking.

Now for SW-L thinking. In this article SW-L considers a recent IMF paper,  the main conclusion of which (to quote SW-L) is that:

“To put it very simply, high debt does impose a burden, because the taxes required to pay the interest on that debt are ‘distortionary’ - for example income taxes prevent people from working as much as they should. But cutting debt also imposes a burden - taxes have to be raised to get debt down. Analogies between households and governments are misleading: while we as individuals do not live forever and therefore need to pay back debt eventually, the state can act as if it will go on forever. We therefore face a trade-off: is it worth paying higher taxes now in order to reduce government debt so we can pay lower taxes in the future?”

I have four objections the above passage as follows.

1. There is the phrase “high debt does impose a burden”.

There is a distinction between debt burden in the sense that there’s a small debt but a high rate of interest is being paid on it, and second, a large debt with a low rate of interest being paid on it.

For reasons give above, there is no excuse for a high rate of interest being paid on the debt, so that is indeed a “burden”. In contrast, there are not bad arguments for “doing a Japan” and having a high debt with a low rate of interest. So even if the bill for interest there is “high”, that can’t necessarily be construed as a “burden”.

2. Taxes are “distortionary”?

On the contrary: one can collect taxes in an almost distortion free way: e.g. a sales tax on everything that consumers buy.

3. “But cutting debt also imposes a burden - taxes have to be raised to get debt down.”

It’s important to make a distinction here between debt which is held domestically (as is largely the case in Japan) and debt held by internationally mobile investor / savers.

To the extent that debt is held domestically, no burden whatever is involved in cutting the debt. That’s because the only good reason to cut the debt is because the private sector seems to have too much PSNFA which is boosting demand by too much. I.e. where PSNFA is excessive, the state should raise taxes (and/or cut public spending) and confiscate some of that PSNFA. That confiscation is not, repeat not, any sort of burden: the sole objective is to cut demand to a non-inflationary level. In fact excess inflation is a real cost to the community, thus dealing with that excess ought to RAISE living standards. Thus ironically, the latter confiscation / tax far from being a burden would actually RAISE living standards. And finally, if interest on the debt is less than inflation, then then the REAL or inflation adjusted rate is negative, which means that the debtor profits at the expense of the creditor: i.e. it's the creditor who bears a "burden".

4. The final sentence of the above passage by SW-L is: “We therefore face a trade-off: is it worth paying higher taxes now in order to reduce government debt so we can pay lower taxes in the future?”

I beg to differ. I.e. there’s no trade off there. Policy should be simply to keep PSNFA at the level that maximises employment without the inflation target being exceeded by too much. That may result in a big rise in the debt this year, followed by an even bigger rise in three years’ time. Alternatively the private sector may have a fit of irrational exuberance in three years’ time, in which case PSNFA might need to be reduced. In short, debt reduction targets do not make sense.

Another SW-L article.

In his second sentence here, SW-L says “If you want to bring the government deficit and debt down, you do so when interest rates are free to counter the impact on aggregate demand.”

The first half of that sentence again implies some sort of debt reduction target. To repeat, there should be no such target.

As to the second half of the sentence, why get into the positon where PSNFA is excessive, and the resulting excess demand has to be dealt with by artificially raised interest rates? I.e. why not, at the first sign of excess demand, just confiscate PSNFA? Ideally interest rates would then not rise at all.

That is not to suggest that interest rates should NEVER be artificially raised. Interest rate adjustment is certainly a useful tool for the state to have. But the AIM should always be, when excess demand appears, to deal with that by confiscating PSNFA (aka run a budget surplus / smaller deficit).

Moreover, there is plenty of evidence that interest rate adjustments don’t work very well. Plus they are distortionary: that is, in the first instance, they influence just the behaviour of those with variable rate loans, and not those with fixed rate loans or no loans at all. That’s a bit like, in the case of helicopter drops, doing drops only on those with red or blonde hair, while everyone else waits from the trickle-down effect of red-heads and blondes going on a spending spree.

A possible objection to the above claim that demand should be adjusted just via changes in the deficit rather than interest rate adjustments is that politicians (particularly in the US) wouldn’t agree to that. (That’s on the assumption that some sort of central bank committee decides the overall size of the deficit, rather than politicians). Well my answer to that is that I’m setting out what I think is the ideal system. If politicians mess that up and put in place a second best system, well it wouldn’t be the first time has happened. (Incidentally, interest rate adjustments are criticised in that work)

As to exactly how a “deficit adjustment only” system would work with politicians retaining a say over strictly political matters while not deciding on the overall SIZE OF the deficit, see this submission to the Independent Commission on Banking. Incidentally while the authors of that work claim the work is dated (and maybe it is dated in some ways), I still think it’s the best and briefest introduction to relevant ideas.

Tinbergen and policy instruments.

Jan Tinbergen said something to the effect that just one policy instrument should be used for each policy objective.  Failure to do that partially explained the 2007/8 crisis and for the following reasons.

In the run up to the crisis it looks like households were putting a more than normal proportion of their resources into property purchases or speculation.  However, aggregate demand was not excessive, and governments to a large extent use interest rate adjustments to control demand. Thus interest rates were not raised. But they SHOULD HAVE BEEN raised given the increased demand for loans by those speculators. That increase in interest rates would have choked off the speculation, if only to a small extent.

Thus there is merit a system where demand is controlled just by varying the deficit with interest rates being left to find their own level.

No comments:

Post a Comment

Post a comment.