Making Markets


Plumbing Matters

The last few days have brought a remarkable, but as of yet unremarked, convergence of attention to the matter of making markets.

We hear about the difficulty of implementing the so-called Volcker Rule, which requires drawing a bright line of some kind between proprietary trading (not allowed) and market making (allowed).

We hear about banks getting out of the market making business, leading to unusually large spreads in corporate bond markets, large enough possibly to tempt others into the business.

We hear about attempts to rein in High Frequency Traders, at least those who make markets in good times but then disappear in bad times.

And of course, behind the scenes, there is the looming role of central banks as dealer of last resort. If the ECB won’t or can’t do it, then why not mobilize the EFSF, suggests the FT.

Time was, most economists thought that investors themselves would provide all the liquidity that markets needed. If price fell just a little bit below value, investors would buy, and if price rose just a little bit above value, investors would sell. Thus, whether you wanted to buy or sell, you could expect to be able to do so quickly, in size, without moving price very much. Efficiency implied liquidity.

No more.

Today we understand that the source of liquidity is the presence of dealers (or their equivalent) who offer trading options in the form of bid and ask prices at which they are willing to buy and sell. The current bible for this point of view is Larry Harris’ Trading and Exchanges: Market Microstructure for Practitioners, but for the deeper economics at stake there is still no substitute for John Hicks’ last book A Market Theory of Money. Taken together, these two books are the best way to explore the fourth tenet of the money view: the importance of dealers for knitting together the different hierarchical layers of money and credit into a unified whole.

This way of thinking tells us that it is essentially impossible to separate market-making from speculation (sorry, Paul). The source of liquidity is the willingness of the dealer to use his own balance sheet to absorb imbalances between supply and demand. That means inventories, and inventories mean risk, both price risk and funding risk, hence speculation.

This way of thinking also tells us that regulatory attempts to prevent banks from taking on these risks, by imposing capital charges or liquidity ratios, will cause market-making to migrate elsewhere in the system, where it is more profitable.

And it says that the problem with HFT is not so much the speed, but rather the question whether the traders are offering trading options, or accepting the offers made by others. The former supplies liquidity, while the latter absorbs it.

Finally, most important, this way of thinking helps us to understand the liquidity dimension of the sovereign debt crisis. In recent days, there has been a push to use the EFSF for bond insurance (see here, and here). The idea is to absorb the first loss, say 20%, but nothing beyond that, and certainly not tail risk.

But this is exactly the opposite of what is needed.

The microstructure literature tells us that the ultimate source of liquidity is the value trader who is willing to buy or sell when the price deviates far enough from value. It is the value trader who creates the “outside spread” that ultimately backstops the ability and willingness of the dealer to quote a much tighter “inside spread”.

In Europe right now, the concern is that private value traders will quote only very wide outside spreads, so that if you need to sell you have to accept a very low price. Fischer Black famously estimated that, even in normal times, the outside spread only keeps price within a factor of two of value. And today the times are not normal.

If the wide outside spread is the problem, then the solution is yourself to quote your own outside spread, inside that of the private value traders. That is what the EFSF could do.

Of course, if the public bid is higher than the private bid, then all sales will be to the EFSF, and it will build up its own inventories, which carry with them price risk and funding risk. That prospect is apparently what is holding everyone up.

But the public value trader is different from the private value trader.

To the extent that price risk for sovereign bonds is really political in origin, the public is the source of that risk and so can internalize it better than the private value trader can. In this way, the solution to the liquidity problem points the way to the solution to the solvency problem.

Furthermore, funding risk is much less of a concern for a public entity whose liabilities would presumably be eligible collateral for discount at the central bank. If expansion of the balance sheet of the EFSF is acceptable QE, then by all means let us do it that way.

I see daylight ahead.

Share your perspective