Sunday, 17 December 2017

Pigou subsidies.


Bit arcane this – so don’t say you weren’t warned.

A Pigou subsidy is an idea I hadn’t come across till Nick Rowe draw the attention of all and sundry to it on Twitter. It’s the idea that a gift, particularly a gift by the rich to the poor, involves a positive externality, i.e. a net benefit for society as a whole. Ergo, the rich should be given a financial inducement to give, in rather the same way as negative externalities like pollution are DISCOURAGED by taxing the polluter.

Nick Rowe’s tweet:




 

 My answer, for what it’s worth, was thus.

Take the simple case of a country divided into two classes, rich and poor. Donations by a rich person to a poor person bring net benefits. Say the benefit is equal (in terms of dollars) to the gift. But to induce a rich person to give, they must be given a Pigou subsidy equal to the amount of the gift.

If the tax needed to fund the Pigou subsidy is raised from the rich, that all nets out to what we already do, i.e. grab tax off the rich in order to subsidise the poor. But if the tax is levied on the poor, they’re no better off.

Doubtless that argument of mine needs tidying up, but I suspect it’s basically valid: i.e. Pigou subsidies are pointless, so we might as well stick to what we already do, i.e. tax the rich and subsidise the poor.


Friday, 15 December 2017

Richard Murphy’s weak grasp of economics.


He advocates a rise in inflation to above the 2% target because that would help debtors*. As he puts it in reference to increased inflation: “….the country could do with some to assist those in debt manage the burdens they face.”

Well now there’s a teensy problem there, namely that creditors and savers are not complete idiots: that is, once they see that an increased percentage of what they save and lend will be eaten away by inflation, they’ll be reluctant to save and lend. Hey presto: interest rates are forced up, which is not exactly of “assistance” to “those in debt”. 

Indeed, that’s exactly what seems to have happened for several years after the 1970s inflationary episode: that is, it took several years of low inflation to convince savers and creditors that inflation had in fact come down PERMANENTLY. Thus debtors were faced with several years of elevated interest rates.

Next, Murphy claims (2nd half of the para in which the above “assistance” point is made) that stagnant incomes are deplorable, and a way out of that is extra demand and inflation.

Well if inflation above the 2% target really does give us extra real GDP, why did no one think of that before? Put another way, the whole point of the 2% target is that economists think that while inflation above that rate clearly gives a faster rate of growth in GDP in nominal terms (i.e. in terms of £s, $s, etc), that is one big illusion. I.e. the consensus in economics is that once inflation rises much above the 2% target, the costs of inflation exceed the benefits. Or put another way, the consensus is that GDP growth is maximised when inflation is around the 2% level.

Of course that consensus might be wrong. It could be that GDP growth is actually maximised when inflation is 4%, 6% or some other figure. But those who want to make that claim need to set out some very detailed reasons to back that claim. 

_________
 

* Article title: “On interest rates, growth and the need for the most massive rethink.”

Wednesday, 13 December 2017

What’s the optimum amount of national debt?


MMTers have solved this one. Others are still floundering, in particular Roger Farmer in this NIESR article on the subject, is all over the place far as I can see  (1). So I’ll run thru this vexed question for the umpteenth time.

First (a point which Farmer does not seem to appreciate) there is little difference between national debt and base money, at low rates of interest, as Martin Wolf and Warren Mosler (founder of MMT) explained. I.e. both are state liabilities, or at least ostensible state liabilities. That’s why MMTers sometimes lump the two together and refer to the pair as “private sector net financial assets” (PSNFA).
 

Thus the first problem in answering the question “What’s the optimum amount of national debt” is that the very concept “national debt” is fuzzy: it has no clear boundaries.

Second, the amount the private sector spends is related to how much PSNFA it has (or if you like, how much “cash” it has). When people win a lottery, their weekly spending rises – a point which is common sense for most people, but a source of much confusion, difficulty and perplexity for some economists.

Ergo, the optimum amount of PSNFA is simply the amount that induces the private sector to spend at a rate that brings full employment. That’s not to say government should instantaneously adjust PSNFA every time there’s a recession. But what it can do is to simply print money and spend it, and/or cut taxes, and that will tend to raise PSNFA. So under that policy, PSNFA (or “the national debt” if you like) will always tend towards its optimum level, a policy advocated by Positive Money.

 

Interest on the debt.

Interest paid on the debt itself influences the “full employment equilibrium” level of that debt: that is, the higher the rate of interest paid on the debt, the more of it the private sector will be willing to hold, all else equal, as Warren Mosler pointed out.

A country which issues its own money can pay any rate of interest it likes on its debt, as MMTers have long pointed out. E.g. if a country thinks the rate is too high, it can reduce it by printing money and buying back debt, and then dealing with any excess inflation by raising taxes and “unprinting” the money collected.

So what’s the optimum rate of interest? Well Milton Friedman and Warran Mosler said “zero”: i.e. they said “don’t pay any interest at all”. And I must say I rather agree: certainly I don’t see any point in paying anything more than the rate of inflation, i.e. I don’t see a reason to pay a positive REAL (i.e. inflation adjusted) rate of interest. That’s for the following reasons.

First, the only really good reason to borrow is if you’re short of cash, and governments are NEVER short of cash: they can print the stuff, plus they can grab near limitless amounts from taxpayers. If a taxi driver wants a new taxi, and happens to have the cash to hand, he won’t borrow in order to buy the taxi. But then taxi drivers probably have more sense than economists.

Second, the old argument that government should borrow to fund infrastructure investment does not make sense because the ENTIRE education budget is an investment, so why don’t we fund that entirely via borrowing?

Third, I set out more argument against government paying any interest on its liabilities here (2).

So, to return to the original question, i.e. what’s the optimum amount of national debt or more properly, PSNFA? The answer is “whatever brings full employment”. And that very much ties up with Keynes’s dictum: “look after unemployment, and the budget looks after itself”. As to exactly what that amount of PSNFA will be as a percentage of GDP, that will depend on how thrifty or spendthrift citizens of the country are. The Japanese are relatively thrifty, thus their optimum PSNFA:GDP ratio will be on the high side. In other countries, the ratio will be lower.


________
Title of Farmer's article: "How much debt do we need? My answer: 70% of GDP".
Title of my article: "Government borrowing is near pointless".

 


Monday, 11 December 2017

The flaw in interest rate adjustments.



The common practice of adjusting interest rates so as to adjust stimulus makes no sense. Reason is thus.

It is widely agreed in economics that the optimum or “GDP maximising” price for anything (including the price of borrowed money) is the free market price, except where it can be shown that the free market is defective or where there is “market failure” to use the jargon. And in the case of the rate of interest, there is no obvious obstruction to the free market working: that is, savers shop around for the bank which offers the best (i.e. highest) rate of interest, and mortgagors shop around to find the bank which offers the best rate of interest from their point of view (i.e. the lowest rate).

The rate of interest is also influenced by the amount government borrows. Unfortunately there is no general agreement as to how much government should borrow. Milton Friedman and Warren Mosler argued that governments should borrow nothing, though Friedman thought there was a case for borrowing in war-time. I argued likewise here. (Title of article: "Government borrowing is near pointless".)

An alternative and popular idea is that government should borrow to fund infrastructure. But a flaw in that idea is that the entire education budget is investment of a sort. So should all education spending be funded via borrowing rather than via tax? There are no easy answers to that, though I argued here a few years ago that (in line with Friedman and Mosler thinking) government borrowing makes little sense.

So in the absence of any totally clear answer to the question as to how much government should borrow, let’s assume the optimum amount to borrow is X% of GDP.

Now let’s assume an economy requires stimulus. One way of imparting stimulus is to simply have the state print money and spend it, and/or cut taxes. And if you think that sounds outlandish, it’s actually not: having government borrow more with the central bank then printing money and buying back government issued bonds (“Gilts” in the UK) comes to the same thing as the above “print and spend” ploy. And the latter “borrow more and then buy back” is exactly what numerous governments have done since the 2007/8 crisis.

Note that that does not alter the above mentioned X%. At least there is no obvious reason why X should change simply because of some stimulus. I.e. a dose of stimulus will raise GDP by some percentage, but if for example there’s an argument for having government borrow to fund infrastructure and nothing else, then the total amount invested in infrastructure will presumably rise pari passu, more or less. Thus borrowing to fund infrastructure will remain at X%.


Interest rate adjustments.

A second way of imparting stimulus is to cut interest rates, and that’s done by having the central bank print money and buy up government bonds. But that reduces the amount of government borrowing to below X%. I.e. the total amount of government borrowing is then less than its optimum or “GDP maximising” level.

Provisional conclusion: stimulus should always be imparted essentially by having the state print money and spend it, and/or cut taxes.


What’s the economy for?

The latter conclusion ties up with a very common sense observation, namely that the basic purpose of the economy is to produce what people want, booth in terms of what they choose to buy out of their disposable income and in terms of the publically provided items (free education  for kids etc) that people vote for at election time.

That is, given a need for stimulus (i.e. assuming the economy is working at less than capacity) the types of spending that need boosting on the basis of the latter “basic purpose” idea, is household spending and public spending. And households and the authorities responsible for public spending can, if they see fit, spend some of that extra money on extra interest to fund more borrowing.

But there is no obvious reason to assume that given a need for stimulus, that it’s JUST borrowing that needs to be boosted.


Friday, 8 December 2017

Random charts - 47.

Large pink text on the charts below was added by me.
















Friday, 1 December 2017

Higher bank capital ratios raise bank lending???



David Miles (economics prof at Imperial College, London) makes the odd claim in the Financial Times that raising bank capital ratios can result in banks lending MORE.

A large majority of those who have examined this question either think that raising capital ratios has no effect on bank lending or that the effect is to CUT lending. Those who back the “no effect” claim normally cite the Modigliani Miller theory, while those who claim that bank lending is cut normally claim that the MM theory is defective. Far as I can see, criticisms of MM are pretty feeble. I run thru them under the heading “Flawed criticisms of Modigliani Miller” here. So I conclude on that basis that raising bank capital ratios has no effect on bank lending.

However, the higher bank capital ratios are, the more difficult it is for commercial banks to create / print money. And as Milton Friedman among others explained, when the ratio is 100% (favoured by Friedman and others), then commercial banks, as Friedman explained, cannot create money at all. Plus the “right to print” pretty obviously amounts to a subsidy of commercial banks (as explained by Joseph Huber (p.31)). And cutting a subsidy for an industry contracts that industry. Thus I would claim that raising capital ratios  will in fact cut lending.

However, given the enormous expansion of lending and debt in recent years, that is hardly a big problem. Plus the deflationary effect of less lending is very easily countered via conventional stimulus, fiscal and/or monetary.



Thursday, 30 November 2017

Increase public investments when interest rates are low?



A popular belief among the great and the good is that since governments can borrow at around a zero real rate of interest, governments should borrow more to fund public investments like infrastructure. E.g. this Oxford economics prof says “The obvious response is for the government to borrow to increase public investment, particularly when it is so cheap to do so.”

Unfortunately there’s a problem with that argument which is that the only reason governments can borrow at ultra-low rates of interest is that governments have coercive powers: they can confiscate money from taxpayers with which to repay their creditors even when their investment projects are a flop.

Thus the rate of interest that should be charged to public investments is the sort of rate that a private infrastructure provider would have to pay, and that is significantly more than the approximately zero real rate of interest that many governments pay.

The original funders of the English-French channel tunnel lost nearly all their money. So what rate of interest would you want for funding a similar project in the future?