Are Market Monetarists mad?

I explained here in February this year that Market Monetarism leads to the state ultimately owning all assets. It seems that Scott Sumner, one of the main proponents of MM has now sort of admitted this.

My conclusion is: God first makes mad….

Fractional reserve banking causes house price increases?

I support Positive Money because PM backs full reserve banking, but some of PM’s other ideas a bit strange, in particular the idea that the existing banking system is responsible for elevated house prices in the UK, a claim PM makes here. There are four problems with that idea as follows.

1. Other countries.

Other countries have EXACTLY THE SAME banking system as the UK, yet while house prices trebled between 1970 and 2002 in the UK in real terms, there was no increase at all in Germany and Switzerland! At least that’s the case according to the chart below which comes from a Policy Exchange study. Incidentally I suspect that “treble” figure is too much, but that doesn’t detract from the basic point which is that if house price increases have been substantially more in the UK that other countries with the same banking system, then it’s a near certainty that it’s not the banking system that explains the relatively large house price increase in the UK.

2. History.

Second, the existing or “fractional reserve” banking system has been going for SEVERAL CENTURIES! There is thus no obvious reason why that banking system has had any more of an effect on house prices in the last ten, twenty or thirty years than the last 110, 120 or 130 years.

3. Supply and demand.

If the banking system DOES PROUDCE a bigger demand for houses than would otherwise be the case, that won’t result in higher house prices if the supply of stuff needed to build houses (land in particular) is elastic. But of course the supply of land is not elastic, as the Policy Exchange article makes clear. Quite the contrary: in the UK the price of land for building is very restricted. So that’s the main explanation for house price increases in the UK. 

4. Money creation.

Third, PM claims “A major cause of the rise was that banks have the ability to create money every time they make a loan.” Now the problem with that argument is that there is no sharp dividing line between money and non-money. In particular, it is widely accepted around the world that so called money in term accounts does not constitute money where the “term” is more than about two months. That is, if it takes the depositor longer than about two months to access to their so called money, then what they possess is not really money. And indeed that is a policy which PM itself goes along with in that it does not count so called money in its investment accounts money where the “term” is sufficiently long.

Now as I explained in section 1.12 here and as others have explained, a bank cannot increase its loans unless there is a corresponding increase in people willing to lend, i.e. unless the number of savers increases. And a mortgage is a LONG TERM loan, thus that loan will TEND TO BE balanced by a LONG TERM DEPOSIT. That of course is not entirely true in that banks engage in maturity transformation, but there is certainly SOME TRUTH in the claim that long term bank loans are balanced by long term deposits. And if those deposits are “term” accounts, then that’s not money for reasons given above, thus  . . . roll of drums . . . it is not entirely true to say that when a bank extends a loan of £X that £X of new money is created. That is, the EXACT AMOUNT of new money created is very debatable.

But it really doesn’t matter how much new money is created. That is, if a bank grants a loan of £Y, and recipients of that new money decide to put the whole lot in two month plus term accounts, the effect  on house prices will be exactly the same as if the savers put their money into current or checking accounts and didn’t spend it. Thus the whole “money creation” point is irrelevant. It has no bearing on the house price boosting effect.

Miller & Modigliani’s critics are clueless.

Under full reserve banking, entities that lend to mortgagors and businesses must be funded just by shareholders. Now that might see a bit extreme: that is, why not just go for the sort of capital ratio advocated by Martin Wolf and Anat Admati namely about 25% rather than the above 100%? 25% is way above the ratio that existed prior to the crunch and it’s way above the sort of ratio advocated by Basel, Dodd-Frank, etc. Doesn’t 25% reduce the chance of a bank / lending entity failing in any given year to about one in a hundred?

Well not in the case of small banks in the US. That is, it’s far from unheard of for the assets of those banks to fall to below 75% of their liabilities. So in those cases, a 25% capital ratio does not save the day. But let’s concentrate on large banks, and let’s assume that a 25% capital ratio reduces the chance of failure in any given year to about one in a hundred.

That raises the question as to whether it is worth going for a 100% capital ratio so as to reduce the chance of failure to about one in a thousand. Well the answer to that all depends on the COSTS of attaining that additional safety: if the costs are negligible, then why not go for the extra safety?

Now a popular argument against raised capital ratios is that bank shareholders demand a higher return than depositors because of the risk run by shareholders, ergo, so the argument goes, the higher the proportion of bank funding that comes from shareholders, the higher the cost of funding the bank, an argument that banks know perfectly well to be a load of nonsense. And the argument is nonsense for reasons set out by Modigliani and Miller (MM).

The MM theory as applied to banks runs essentially as follows. The risks involved in funding a given bank are a GIVEN. Thus if the proportion of funding that comes from shareholders is raised, the total cost of covering the latter risk remains unaltered: i.e. all that happens is that the risk per share or per shareholder declines. Thus raising the proportion of bank funding that comes from shareholders has no effect on the cost of funding the bank.

Moreover, numerous non-bank corporations are funded mainly by equity rather than by debt (short or long term). If funding mainly via equity really was expensive, those corporations would have cottoned onto the fact by now.

The MM theory HAS BEEN criticised however. These criticisms are actually a motley collection of half-baked nonsense. But anyway, let’s run through them.

Incidentally, raised capital ratios WILL RAISE the cost of funding banks in that those raised ratios necessarily reduce the extent of taxpayer funded bank subsidies. Put another way, there is a big reduction in the likelihood of taxpayer funded bank bailouts being needed when capital ratios are raised from about 3% to about 25%. While there is a much smaller reduction in the chance of bailouts being needed when ratios rise from 25% to 100%.

But there is nothing wrong with disposing of bank subsidies. Anyway, let’s get stuck into the half-baked criticisms that have been made of MM.

1. Tax. 

About the most popular criticism of MM has to do with the different tax treatment of bank capital and bank debt. At least that tax point is the first criticism cited by Douglas Elliot, David Miles, and Vickers, and it is only one of two criticisms cited by Lev Ratnovski and the ONLY criticism cited by Anil Kashyap. Urs Birchler also cites the tax point.

This tax criticism of MM is simply that if the tax treatment of bank capital and bank debt is different, then MM does not work out in the real world in the same way as it does in theory.

The very simple answer to that is that tax is an entirely ARTIFICIAL imposition. Thus for the purposes of gauging REAL costs and benefits, tax should be ignored. If a big tax was imposed on apples, and for no good reason, that would not mean that the REAL COST of producing apples had risen.

A fifteen year old of average intelligence ought to be able to work that out.

It is INCREDIBLE that the above six so called “authorities” who cite the tax criticism of MM have apparently not tumbled to the latter simple flaw in the tax criticism, but that seems to be the case. And if the latter tax criticism is the best that the critics of MM can do, then rest assured that the other criticisms (dealt with below) are near hopeless.

2. Different returns on capital and debt.

Ratnovski’s only other criticism of MM, which he does not present with what seems a lot of confidence, is that assuming the return on bank capital is 15% and the return on bank debt is 5%, then the more capital there is, the higher the cost of funding the bank.

Answer. Well of course, but it’s PRECISELY the latter sort of 15%/5% assumption that MM demolishes. Ratnovski’s point there is a bit like saying “if the Earth was flat we would not have weather satellites”. The answer is  . . . wait for it . . . that the Earth is actually round, so we do have weather satellites.

3. Incorrectly priced deposit insurance.

The second criticism of MM made by Miles (also the second criticism made by Elliot is that the charge made for deposit insurance may not reflect the risk, in which case MM would not work out in the real world the way it does in theory. 

Answer. The flaw in that argument is much the same as the flaw in the above “tax” criticism: for the purposes of gauging REAL costs and benefits, any “incorrect” or artificial charges should be ignored. That is, in such cost / benefit calculations or arguments, CORRECT OR ACCURATE charges should be assumed, even if those are not the charges that obtain in the real world.

4. Asymmetric information.

Elliot’s third criticism (also made by Birchler) is that new share issues have to be offered at a discount because those buying shares do not have perfect information about what a bank does.

Answer. Exactly the same applies to bonds issued by a bank and to deposits. That is, depositors and bond holders do not trust banks 100%, thus depositors and bond holders “lend” less to banks, and charge a higher rate of interest for doing so than if they had perfect information about the bank.

5. Regulators and banks do not agree on MM.

Birchler seems to think that the fact that banks and regulators do not agree on the relevance of MM is a weakness in MM. As he puts it in reference to MM, “Bankers and regulators thus fail to agree on the relevance of even the validity of a half-century old theorem.” (The “theorem” is of course MM).

Answer. This might be news for Birchler, but cops and robbers often disagree! In fact it is not unknown for them to employ extreme violence against each other. And if there is any doubt that banks are robbers, remember that they have been fined around $100bn in the US recently. Yes that’s billion, not million.

6. Bank shareholders want leverage.

Another feeble point made by Birchler is that bank shareholders want leverage. As he puts it, “Firstly, once a bank is already leveraged, shareholders are tempted to push leverage further leverage”.

Answer. Well of course they will. Pickpockets, if they were as brazen as bankers, would be “tempted to push for” the right to pick pockets.

7. State guaranteed deposits.

Birchler’s next criticism of MM is that banks do not need more capital because they can fund themselves from taxpayer backed deposits. As he puts it “..banks can raise insured deposits (or liabilities with implicit state guarantee).”

Answer. Well of course, but taxpayer backed deposits are a subsidy of banking! Hopefully readers will forgive me if I ignore the rest of Birchler’s article. It is clearly a waste of ink and paper.

8. Banks will increase risks.

Elliot’s fourth criticism is that higher capital ratios, improve bank safety, which may induce banks to take bigger risks.

Answer. The WHOLE POINT of raising capital ratios is to prevent banks offloading risk onto taxpayers. And if banks find they cannot unload that risk, i.e. have to carry the risk themselves, they are almost bound to become more cautious or responsible, rather than (as suggested by Elliot) take bigger risks.

If the state guaranteed to replace my car if I write it off in an accident, that is an inducement to irresponsibility on my part. If I then have to insure it myself, assuming there is a significant no claims bonus, then I will drive more responsibly.

But if by any chance higher capital ratios DID RESULT in increased risks, why should we care? As long as the risk is taken by shareholders rather than taxpayers, then that is free markets working in the way they should. Oil companies take big risks when drilling for new oil deposits: sometimes there is no oil there. That is how capitalism and free markets work.

The beauty of a bank that is funded just by shareholders is that if risks do not pay off, it cannot go insolvent, which disposes of systemic risks. As George Selgin put it in his book on banking, Selgin (1988), “For a balance sheet without debt liabilities, insolvency is ruled out…”.

9. The transition to full reserve.

Elliot’s fifth criticism is that the transition to higher capital ratios may involve problems, and that banks may react to higher capital requirements by cutting lending rather than actually acquiring more lending.

Answer.  The “transition” point is a bit feeble. Granted there may be transition costs, but if higher capital ratios are actually the way to go, then the country will reap the benefits of that for the next hundred or two hundred years, which will make any transition costs irrelevant.

As for the possibility that banks will react to higher capital requirements at least to some extent by cutting lending, that is not a possibility: it’s a certainty. Reason is that higher capital ratios reduce bank subsidies  which in turn raises interest rates which cuts total debts and lending.

Elliot claims that “If banks do cut back on credit provision, then either the economy is likely to be slowed down, or less regulated entities will pick up the lending slack, bringing up other risks that will be covered in the next section.” The answer to that is that higher capital requirements involve removing or reducing subsidies and subsidies misallocate resources, i.e. reduce GDP. Thus far from less lending causing a REDUCTION in GDP, as claimed by Elliot, the actual effect would be to INCREASE it.

10. Shadow banks.

Elliot’s final criticism is that increased capital requirements will drive business to the shadow bank sector. The answer to that was given by Adair Turner former head of the UK’s Financial Services Authority. As Turner said, : "If it looks like a bank and quacks like a bank, it has got to be subject to bank-like safe-guards." I.e. all banks, or at least all banks above some minimum size should abide by the same rules.

Regulating one lot of banks, but not another lot makes as much sense as forcing males to abide by speed limits on the roads while not forcing females to do so (or vice versa).

Conclusion.  

There may be some flaw in MM that I have not spotted. But I’ve been through a reasonable selection of MM critics, and their criticisms are frankly pathetic.

Game set and match to MM far as I can see.

Unimpressive paper by Bill Mitchell and Martin Watts on JG.

Job Guarantee (JG) is simply a new name for an idea that has been around for centuries, namely the idea that government can act as employer of last resort. The idea was put into effect, for example in the US in the 1930s in the guise of the “Work Project Administration”. The idea was even put into effect in Ancient Athens 2,500 years ago.

Anyway, Bill Mitchell and co-author produced a paper on this subject at the end of last year.  The first odd aspect of this paper is that it uses the term “NAIRU” in the conventional sense, i.e. the sense employed in dictionaries of economics or economics text books.

That’s in stark contrast to the bizarre sense in which Bill Mitchell normally uses the term NAIRU. That’s something like “a wicked evil plot by employers or neo-liberals to do down the working classes”. I may be exaggerating a little or taking the p*ss just a little there, but if you don’t believe me, then Google NAIRU and “Billyblog”, which is Bill Mitchell’s blog. You’ll see what I mean.

Anyway, I’m all in favour of sticking to dictionary definitions, so Bill’s use of “NAIRU” in the conventional sense is welcome. Moving on….

The basic argument in this paper is as follows.

1. Any economy faces capacity constraints. The authors of the above paper go into a lot of unnecessary detail on the question as to whether after a recession, capacity is permanently reduced, i.e. whether there is a hysteresis effect that stems from recessions. And that effect, so the authors claim, can come because of two factors: first a permanent deterioration in skills, and second, inadequate investment (due to the recession) by employers.

The existence or otherwise of that post-recession capacity reduction is actually irrelevant to the subsequent argument, as will hopefully become obvious.

2. The authors then try (in section 4.3) to deal with a criticism of JG (made amongst others by myself) namely that unless JG type work is going to be hopelessly inefficient, such schemes have to employ some capital equipment, materials and have some permanent skilled labour as well as the relatively unskilled labour which is what JG employees tend to be.

Now there is a big problem there as follows. Assuming the economy is already at capacity, government cannot simply place orders for extra capital equipment and materials: if it does, aggregate demand (AD) will rise to above the “capacity” or “NAIRU” level, and inflation ensues. Worse still, JG withdraws permanent skilled labour from the regular workforce. Thus to get JG going, AD has to be raised to above the capacity level, meanwhile aggregate supply has been reduced because permanent skilled labour has been removed from the regular workforce. Excess inflation is bound to ensue.

The authors’ answer to the latter dilemma is plain bizarre. They say, “However the implementation of the JG exploits the spending capacity of a currency - issuing  government, which is not constrained by expectations of future aggregate demand . This stands in contradistinction to the spending decisions of private firms that are guided by profitability considerations and constrained by endemic uncertainty. In other words, the JG creates its own productive capacity each time it takes on a new worker.”

Now let’s run thru that passage, starting with “JG exploits the spending capacity of a currency - issuing government, which is not constrained by expectations of future aggregate demand..”

Well it’s not just JG that exploits or can exploit the “spending capacity of a currency issuing government”: dozens of economists and groups of economists with no interest in JG believe in “exploiting” that spending capacity. Keynes did. Milton Friedman did. Present day advocates of full reserve banking like Positive Money do.

In short, the fact that governments can print money and spend it, is irrelevant to the issue here. Put another way, we don’t need JG in order to implement the above “exploitation”, which is what the authors seem to suggest.

Next, there’s the bit about “spending decisions of private firms being constrained by endemic uncertainty”. So AD is constrained by that “uncertainty”? Is that what the authors are saying? If so, how come AD is sometimes excessive and causes inflation?

In short, the whole idea that when an economy is at capacity, there won’t be an inflationary effect when government prints money and orders additional capital equipment and materials for JG schemes is just pie in the sky.

To elaborate on that, when the economy is at capacity, firms JUST CANNOT produce more: reason is that they face labour shortages, shortages of skilled labour in particular. If additional demand DOES APPEAR (and it makes no difference where that demand comes from) inflation will rise. The extra orders may come from other countries, the orders may be from government trying to set up a JG scheme: it makes no difference.

As to the hysteresis effect mentioned above, that’s irrelevant. If that effect IS OF significant proportions it just means NAIRU is a bit higher than if the hysteresis effect is of negligible proportions. That does NOT AFFECT the basic theoretical arguments dealt with just above.

So is JG scuppered?

Well no. The reasons briefly are as follows.

JG clearly works and the reason why is best illustrated by considering an ultra-simple or crude JG scheme as follows. Government simply tells the unemployed that their benefits are henceforth conditional on walking up and down their street keep it free of litter. Hey presto: unemployment vanishes.

Of course the latter JG jobs are very unproductive. But forget that. The important question is WHY DOES THAT JG SCHEME WORK (admittedly in a strange sense of the word “work” )? Reason is that there is no increase in AD and no reduction in aggregate supply (because the JG people will still seek normal employment). Indeed, given the unpleasant nature of the work, they’ll probably seek normal jobs MORE ENTHUSIASTICALLY than when offered unemployment benefit.

Now let’s move on from the latter near ridiculous form of JG work to JG work which involves some capital equipment and permanent skilled labour. How to supply JG with the latter factors of production? The answer is just nick them from the private sector, rather than increase TOTAL DEMAND for capital equipment and materials. That is: cut AD and reinstate that AD in the form of orders from government for capital equipment and materials.

That will certainly reduce output in the private sector, but taking the economy as a whole, output of capital equipment and materials remains constant, while the total number of people employed will rise because of the extra employees in the form of JG employees. Ergo GDP ought to rise.

But that’s NOWHERE NEAR the end of the argument. For example there is the question as to whether JG employment should take the form of what might be called “specially set up” employers, which is what the above authors seem to envisage and which is also largely the form that the WPA in the US in the 1930s took, or whether JG employees should be placed with EXISTING employees. The answer is the latter for reasons I spelled out here.

There is also the question as to whether JG employees should do just public sector type work, or whether JG employees should be subsidised into work with PRIVATE SECTOR employers. The answer again is the latter. See same paper.