The trigger for the next financial crisis

We know that financial crisis are recurring. The last one was in 2008, almost 8 years ago, so it is not difficult to forecast that a new one is in the making.

The trouble for investors is that it is almost impossible to forecast when a new crisis will arrive. It may be next month, next year or in five years. Who knows?

Another annoying feature of financial crisis is that it is equally difficult to foresee which drop will spill the glass, that is, what will be blamed as its trigger.

Paul Krugman just listed two possible candidates – China and oil – in his NYT op-ed “The 8 A. M. Call”.

So, let me add another candidate – IRRBB Interest rate risk in the banking book – caused by the excessive reliance of banks on the derivatives market.

This choice may sound odd, given that the Basle Committee has been working hard on new standards for this risk which will be implemented in 2018, aimed at preventing IRRBB to materialize when interest rates normalise from historically low levels. Indeed, the notional value OTC derivatives which at end-2014 was 6.5 times larger than all outstanding debt securities has been reduced by 70 trillion USD in the first half of 2015.

It really does not matter if the crisis is triggered in these markets, or any other place like the M&A or the currency markets. The “fire” can be started anywhere, as long as the “wood” is close enough to spread it by contagion.

What matters is to know whether the “firemen” are capable of containing it.

In this regard the perspectives are frightening, given the resistance to apply the so-called Volcker rule to ring-fence commercial banking from investment banking, the exhaustion of ammunition by central banks to deal with a liquidity trap and the balance of public finances in some major economies.

Securitization: A good idea gone bad

Unbundling is usually a good idea. It gives the consumer a choice between price and quality. For instance, if you go to a store to buy apples would you prefer them in a single box at a single price or would you favour separate boxes with different prices for categories I, II and III? If you have an average size budget, can pick them on your own and are one of the first customers in the store you would certainly prefer them bundled because you could pick the top quality and pay the average price. On the contrary, if you were in a low or high budget and arrived later you would prefer them unbundled in three different boxes. The vendor would face the same dilemma when buying from the farmer.

In general, since most buyers cannot be the first to arrive at the shop, the unbundled solution is better for both buyers and sellers. Yet, one must question if this principle is valid in the case of perishable or dangerous products (e.g. risky financial products like sub-prime mortgages or auto loans).

Judging from the recent rebound in securitization it seems that the answer is yes. That is, the buyers (investors) prefer to pick between various categories of risk instead of buying a single security sold by the vendors (banks). However, an eventual resumption to pre-2008 levels might create a number of potential perils.

First, buyers (investors) may be poorly diversified due to a temptation to buy from the low priced fruit boxes (i.e. high yield tranches).

Second, because financial products are more difficult to classify than fruit, buyers (investors) need the opinion of outside experts (rating agencies) which, being procured and paid by the vendors (banks), are prone to taint their ratings.

Third, vendors (banks) can select perishable fruits to sell on outside stalls (sell loans to special purpose vehicles) and save store space (save on capital) that can be used to serve more profitable clients (provide more loans). Yet these extra profits tempt vendors (banks) to aggressively promote the sale of perishable fruits (lower quality loans).

Fourth, to meet this increase in demand vendors (banks) widen their supply network to lower quality farmers (loan originators).

Finally, and most importantly, vendors (banks) give up on buying unbundled fruit (loans) from farmers (loan originators) and the less scrupulous suppliers will begin mixing up rotten fruit (subprime borrowers) in the containers (loan portfolios).

Individually considered each one of these risks may not be enough to offset the benefits of the outside sale of unbundled fruits (securitized issues). However, in aggregate the risks may be compounded to the point of creating systemic risk. The greatest danger is that banks give up their key function of screening creditworthy borrowers to become commission-driven businesses.

So, just as planning regulations limit the portion of walkways that can be taken by outside stalls, banking supervisors should establish limits on how much debt banks can offload through securitization.

Such limits would not constitute market restricting policies. Instead, if designed properly, they would be an important market-perfecting tool for the development of market capitalism in the financial sector.

Mankiw's proposal for contingent convertible debt

In the ongoing debate on financial regulation, most reform proponents seem to be under the illusion that it is possible to prevent future financial crisis. Therefore, Mankiw’s article on Trying to Tame the Unknowable is a welcome alert.

Proposals to prevent future taxpayer bailouts for financial institutions revolve around three key ideas:
1) Limiting the type of activities that banks can do (e.g. the Volcker rule on proprietary trading);
2) Capping the size of banks considered too big to fail; and
3) Requiring banks to have more capital to cope with the higher levels of leverage used today (e.g. Greenspan’s proposal to go beyond the current Basle II).

Mankiw’s favorite proposal is to require banks, and perhaps a broad class of financial institutions, to sell contingent debt that can be converted to equity when a regulator deems that these institutions have insufficient capital. This debt would be a form of preplanned recapitalization in the event of a financial crisis, and the infusion of capital would be with private, rather than taxpayer, funds. Think of it as crisis insurance.

This is an interesting idea on how to increase bank capitalization while privatizing the risk of reckless lending. However, its fundamental weakness is that rating agencies and regulators can hardly be trusted to decree such debt conversion before losses are too big to be absorbed by this share of capital alone.

The one thing that the history of speculation teaches us is that nobody is willing to remove the bowl of punch while the party is still going on. For instance, this last year alone, with inflation below 3%, the stock market has risen more than 40%. Would anyone be willing to classify this as a bubble and to force banks to stop lending to finance stock purchases? I doubt.

The only way the regulation could work would be to set up pre-defined conversion rules based on asset inflation targets. But, which assets classes? Shall we use rules based on price indexes for stocks, real estate, fixed-income, or commodities? And what should be considered a speculative price run over, say three months? 20%, 40% or 60%? These questions as well as the size of this quasi-capital buffer would raise many interesting debates.

However, the rules would only work if we could agree on a simple set of pre-defined conversion rules. If we are sufficiently naïve or optimistic to believe that such agreement is possible, let us advocate the Mankiw’s rule as a good policy.