Andrew Haldane: Financial arms races

Happy birthday, Aaron.

One thing I know for sure from my DPD graph is that the ratio varies: the amount of debt there is, per dollar of circulating money, varies. Since I first became convinced of the significance of that graph, I have been expecting to see policymakers turn things around and bring the ratio down.

Well, since the crisis the ratio has been coming down. But policymakers had nothing to do with that.

No, that's not entirely true. Policymakers did create the crisis, after all.


In an anonymous comment some days back (while I was up to my eyeballs writing the 'debt and inflation' posts) Nonny wrote:

OT , but since we were discussing Phillipon's papers on our bloated financial sector a while back , I thought you guys might like this related paper by Andrew Haldane :

http://www.bankofengland.co.uk/publications/Documents/speeches/2012/speech565.pdf

It covers the same ground as in his recent INET presentation ( i.e. the futility and resulting destruction of the "arms race" in finance ) , and he's as good a writer as he is a speaker.

Well I must say Nonny is right: Haldane is an excellent writer.

Maybe too good. I'd put him in a class with Krugman. With both of them, I worry that I'll be convinced by the good writing when what I really need is good economics.


Setting aside the quality of writing, there is still much of value in Haldane's paper, I think.

Page 4:

By any historical metric, the pre-crisis period was an extra-ordinary one for the financial sector. This is no better illustrated than in the returns to financial sector labour and capital. Between 1989 and 2007, the nominal gross value added of UK financial intermediaries rose more than threefold, or by about 8% on average per year. Over the same period, value-added of the non-financial corporate sector rose by about 5½% per year.

Assuming they start out equal and grow at the rates Haldane reports for the 1989-2007 period, finance grows from 100 to 400 while the productive sector grows from 100 to 262.

Finance increased 52% more than the productive sector, by Haldane's numbers.

Page 5:

Measures of executive compensation rose even more dramatically...

In 1980, a similarly-skilled individual earned about the same in investment banking as in the non-financial sector. By 2005, that same person would be benefitting from a salary four times that of their non-financial sector counterpart.

What explains this extra-ordinary period of high returns in finance? Arithmetically, the answer is simple – leverage.

Leverage. Debt and more debt. But the simple answer is not good enough for Andrew Haldane.

"But what drove those behaviours in the first place?" he asks. Competition, he answers. Competition for "high returns on equity". Competition "in financial sector pay".

As wages and equity returns ascended, the races in pay and returns joined forces. That spiral defined the pre-crisis arms race in financial returns. It generated an equilibrium of synchronously high returns and pay, as banks unilaterally militarised as defence against their competitors.

The result, Haldane says, was "sub-optimally risky".

Sure. But is Andrew Haldane suggesting that the problem is in our competitive spirit? That the problem is human nature? One must wonder then why it becomes a crisis only occasionally.

And if Andrew Haldane has doubts about competition, then we may have doubts about Andrew Haldane.

But I don't see it. There was competition in the non-financial sector too, without the extreme effect it had in finance. What finance did have, that nobody else did, was demand for the product. Demand for loans.


There is more. Then Haldane offers some interesting conclusions:

What public policy lessons might be drawn from these financial arms races? First, because they generate sub-optimal outcomes, policy intervention can be justified.

Second, to be effective this intervention needs to constrain behaviour across the financial system as a whole...

Third, what tools might such agencies bring to bear in tackling financial arms races in return, speed and safety? With hindsight, it is not difficult to identify instruments which might have slowed the pre-crisis race for excessive returns in banking. The most direct and effective would have been to place tighter constraints on banks’ leverage. This would have defused the return on equity race at source.

Yes, very nice. But how would you design the constraints on bank leverage? How would you design those constraints to withstand the continuous pressures driving financial expansion? What are those pressures, and why does Haldane not address them?

I say we can constrain bank leverage by cutting demand in half. Cut the demand for loans in half. Cut the borrowing in half. Undermine the source of financial profits.

Yes. But if it is so easy, why haven't we done it?

Good question. We haven't done it because it requires changes in the assumptions that underlie policy. But I'm getting ahead of myself.

What we need to do, in order to reduce the demand for loans, is to move away from credit-use, and move toward increased reliance on the dollar. We have to provide the economy with a lot more money that gets turned into a lot less debt. No, not "we". Policy has to provide the economy with a lot more money that gets turned into a lot less debt. Only policy can do it.

And that's where the changes to our assumptions come into play.

The Federal Reserve seems to think that a dollar of credit-in-use is just as good as a dollar of money in circulation. It isn't just as good. It's better (for the lender) and worse (for the borrower) because of the cost of interest that applies to credit but not to money.

Better for the lender and worse for the borrower. Doesn't that explain the growth of finance? Doesn't it explain the laggard performance of the economy?

We need to take income out of the non-productive sector and put income into the productive sector again, where it was when our economy was good.

To policymakers I say this:

 • We need to stop assuming that credit is as good as money.
 • We need to stop encouraging the use of credit to stimulate growth.
 • We need to stop taking money out if circulation to fight inflation.
 • We need to accelerate the repayment of debt to fight inflation.

We need to assure that there is enough money in circulation to sustain the existing level of economic activity, and then use credit for growth. Then, when the economy grows, we need to pay down that debt and counterbalance the deflationary effect by increasing the quantity of money in circulation.

Haldane's Recommendations

In the last few pages of his PDF, Haldane offers several parts of a solution. He wants "tighter constraints on banks’ leverage." He wants "to require banks to adopt [better] performance metrics". He wants banks to avoid "linking remuneration to return on equity targets". He wants to restrict "cash distributions by banks to shareholders and staff".

He wants "to set remuneration codes... for example, maximum ratios of cash distribution and minimum periods of pay deferral." He wants to reconsider "the instruments used for remuneration." He wants "to place direct restrictions on excessive message traffic". He wants "to strengthen market-making commitments." He wants to strengthen "circuit-breakers across exchanges". And he wants "greater transparency".

Andrew Haldane thinks like a banker. He looks for ways to minimize the problems arising from the excessive expansion of finance, but he fails to consider simply reducing the size of finance and reducing the role of credit.


I don't object to Haldane's recommendations. Heck, I don't even know what most of them are. But let's say Haldane is right. Let's say the crisis arose because we didn't have his recommended policies in place.

If we didn't have those policies in place, it is because of policymakers' faulty assumptions about money and credit.

And if you have any doubt about policymakers' assumptions, ask yourself why, at the onset of the crisis, we had $3.50 of credit in use for every dollar's worth of GDP, but only 10 cents of spending money.