There is never a right time to tackle moral hazard…

Ricardo Caballero’s argument in his Financial Times column of July 14,  Moral hazard misconception about moral hazard, is essentially that doing the right thing to minimize moral hazard would be too costly in terms of the likely negative impact of such actions on the real economy.  The way he presents his case is a textbook example of how a combination of lack of commitment/opportunistic behaviour, myopia and strategic interaction between the private sector and the government can create a very bad equilibrium.  I will refute his argument, focusing mainly on the case against bailing out Fannie Mae and Freddie Mac, unless this involves the euthanisia of the existing shareholders of the two GSEs and a material haircut for their creditors.

In what follows I show, first, that even if we wish to keep Fannie and Freddie in their current form, the immediate crisis need not get worse if their shareholders and creditors are treated harshly, thus maintaining incentives for future responsible lending, borrowing and investing.  Second, I show that a more efficient and equitable solution is available that ends the institutional obfuscation inherent in Fannie’s and Freddie’s current form: public sector sheep dressed in private sector wolf’s clothing.

In 1968, the US federal government privatises a public sector entity, Fannie Mae, that subsidises residential mortgage financing for middle America. In 1970 Freddie Mac is created by the government on the same model as Fannie to turn the de-facto monopoly into a duopoly.  The private sector knows that, even though the two GSEs (henceforth F2) are notionally private, with private shareholders and private other sources of finance, their liabilities remain de facto an obligation of the US federal government.  George W. recently confirmed in public that there was an implicit public guarantee for Fannie’s and Freddie’s debt.  Would be creditors to F2 therefore don’t worry about default risk and lend thoughtlessly and recklessly to these institutions. The management of F2 know this too and expand recklessly. The rents from the implicit public guarantee are shared between the mortgage borrowers, F2‘s initial shareholders and the management of F2.

Then stuff happens. The subprime crisis erupts.  F2 contributed only indirectly to the subprime mess, by hogging the middle segment of the home loan market, leaving the real private home lenders just the two tail ends of the distribution: subprime and Alt-A at one end and jumbo mortgages at the other.  More generally, six years of global underpricing of credit risk – indeed of all forms of risk – come home to roost.  A couple of decades of global competitive deregulation of financial markets and financial businesses encouraged reckless lending and borrowing in the advanced industrial countries of the North Atlantic area.  Much of this high-risk lending and borrowing goes pear-shaped. This credit boom, bubble and bust causes a contraction of demand in North Atlantic economic activity at the same time that a global boom in commodities (especially fuel and food crops) contributes to an adverse supply shock and a spike in inflation.

The question arises then as to how to deal with systemically important private financial businesses that have been rendered either illiquid but not insolvent or both illiquid and insolvent.  Systemically important means too large or too interconnected to fail.  F2 , with about 50% of the face value of the stock of US residential mortgages either on their books or covered by its guarantees, are clearly in the too large to fail category.  This has to be done, however, in a way that does not recreate or even exacerbate the moral-hazard inducing distorted incentives that caused the financial crisis in the first place.  The solution for F2 ought to be obvious.

They should never have been privatised (in the case of Fannie) or created as a private corporation sponsored by the federal government (in the case of Freddie).  If subsidisation of home ownership by the tax payer is deemed socially desirable, let it be explicitly funded by the federal budget.  To use these two PSEs as off-budget and off-balance sheet quasi-fiscal agents of the federal government is deceptive and limits accountability for the proper use of government funds.  This crisis offers a wonderful opportunity for ending this anomaly either by taking Fannie and Freddie into public ownership, or by properly privatising the provision of residential mortgage financing.  This can be done in a number of ways that does not involve the need for a capital increase for F2 that depends on attracting private equity capital.  The existing shareholders can therefore be led to the slaughter.  The existing creditors can be given a haircut.  Moral hazard will be minimised.  The US mortgage market will not be harmed.  Specifically, I would proposed one of the two following options:

(a)   Nationalise Fannie and Freddie.  This should be done without giving the current shareholders more than they would have received without the benefit of an implicit or explicit guarantee of public sector support.  That ought to mean that the current shareholders get nothing.  There also is no reason why the other creditors of F2 should be made whole (that is, have their claims on F2 paid in full).  A non-trivial haircut for all creditors, with the size of the haircut varying inversely with the degree of seniority of the unsecured creditors.  If the creditors balk at this, they can be threatened with option (b).  There will be no immediate impact on the government budget, although both the (illiquid) assets and the liabilities of the federal government would rise by a few trillion US dollars.

(b)   Let the Treasury buy the assets of F2 (its mortgage book, its mortgage guarantees, and perhaps part of its portfolio of financial derivatives).  This should be done at an appropriate (and I would say sizeable) discount to their face value or notional value.  The existing unsecured creditors would distributed the proceeds from the purchase of the asset book (somewhere between 2.5 and 3 trillion US dollars probably) among themselves in order of seniority. The existing equity holders would be last in the line of claimants.  My best guess is that the existing shareholders would get nothing and that the other creditors would take a haircut. The federal debt would increase by the 2.5 -3 trillion US dollars required for the asset purchase.  This would do no more than make explicit and visible the implicit and barely visible implicit federal guarantee of F2.

Under both option (a) and (b), the Federal government can either continue a Federal role in the provision of mortgage financing (on any scale) or end this federal intervention in the residential mortgage market.  If there is felt to be a need for a federally owned and federally funded supplier of residential mortgages and mortgage guarantees, the nationalised Fannie and Freddie of option (a) could do the job.  It probably would make sense to consolidate them in this case.  If the federal government decided the case for the state being in the mortgage lending business may not be as strong as once thought, the two GSEs could be scaled down and wound up under option (a).  Under option (b), it could simply sell the assets it acquired to private mortgage market parties.  Unconstrained by liquidity or by capital requirements, the federal government could sell off the mortgage books and guarantees of F2 as market conditions improved.  With a bit of luck, it could make some money for the US tax payer.

Ricardo Caballero objects to letting the existing shareholders of F2 get hammered, because his only conceivable future for US residential mortgage financing involves a continued significant role of F2 in their current form, as private corporations.  Hammering existing shareholders would impair the capacity of F2 to attract new private equity in the future.  Making existing creditors take a sizeable haircut would impair the ability of F2 to attract future external credit.  My proposed solution removes the need for future private equity subscriptions and for access to future external credit by a notionally private entity in the private markets.

The Caballero approach, which I will paraphrase as: let’s bail them out before something truly awful happens to the housing market and the real economy; now is not the time to worry about moral hazard is twice wrong.  It is wrong first because even if we had no option but to continue for the foreseeable future with F2 in their current form and size, his is what macroeconomists call a ‘time-consistent’ solution and game theorists call a ‘subgame-perfect’ solution.  It is however, a dreadful solution with will see the reckless lending and borrowing that created the current crisis replicated and augmented in the future.  Other time-consistent solutions are ignored by Caballero only because of the extreme myopia he attributes to the authorities.  The authorities see the risk and dangers (not least tot their electoral prospects) of the immediate financial crisis and economic downturn.  The future financial booms, bubbles and busts whose incidence and severity will be amplified by baling out F2 without taking the existing shareholders and creditors into the basement and beating them with a rubber hose, are ignored or heavily discounted.  With less extreme myopia, a strategy that involves punishment of shareholders and creditors can be part of a time-consistent equilibrium.

The second reason Caballero’s approach is wrong is, as I argued earlier, that we can save what’s left of the US home finance market without F2  in their current form, and indeed without any need for F2 and without any need for future private equity injections and private credit provision for strange quasi-private or crypto-public GSEs.

Subsidize owner-occupancy if you have to; don’t subsidize residential mortgage borrowing.

Should the US government continue to subsidise borrowing secured by residential property?  I can see no earthly efficiency or fairness argument for this.  There may be externalities associated with home ownership, specifically with owner-occupancy.  In that case, the government should subsidise owner-occupancy, regardless of how the property in question was financed.  This could be done, for instance, by paying the head of an owner-occupying household a fixed amount of money each year, or to give him/her a fixed deduction from property taxes.  But if owner-occupancy is the source of the positive externalities (an owner-occupier will look after his/her property better; this will raise the value of the neighbour’s property and may lower vandalism and yobbism or what not) then an owner-occupier should be subsidized regardless of whether (s)he inherits the property, purchases it outright or borrows against the value of the property.  What is the positive externality associated with household borrowing secured against residential property per se?  Is using your home as collateral a uniquely virtuous act in need of official encouragement?  Would Americans borrow too little (against their homes or across the board?) without these subsidies? Let’s end the nonsense of subsidising this form of borrowing favoured by the middle class.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website