Financial Stability: Retrospect and Prospect

“Financial Stability: Retrospect and Prospect”
Remarks at the Stanford Institute for Economic Policy Research
Lawrence H. Summers
March 12, 2010

I am very glad to have this opportunity to be back at SIEPR and to be back at a time when SIEPR is rightfully celebrating its past accomplishments and looking forward to the contributions that it’s going to make to economic policy in the years ahead.

I divide public policy problems into two categories. 

Some – like Middle East peace, the budget deficit, Social Security – are profoundly difficult problems, but primarily they are profoundly difficult problems because while everybody can sort of see what the solution is, it is enormously difficult to accomplish politically.

There’s a second class of public policy problems that includes aspects of forging international agreements on global climate change and that includes perhaps certain aspects of the very difficult tradeoffs in health care costs.  That involves what’s going to be my subject tonight – financial stability – where, in addition to formidable political challenges, there are formidable intellectual challenges in finding the right path forward.

But first, a bit of a progress report on the issues I talked about two years ago.

I was right about some things.  I was wrong in not assessing and looking forward to the full extent of the liquidity problems that befell the American economy.  I could not have imagined that the General Electric Corporation would be borrowing money overnight because it was unable to borrow money for a one-week time span.  I could not have imagined that a profoundly conservative and free-market administration would invest several hundred billion dollars in private financial institutions because it saw no choice.  And I could not have imagined that it would prove necessary for the federal government to guarantee literally trillions of dollars in money market funds and in other places.

The extent and virulence of the liquidity panic was not something that I foresaw then, and until the very moment it arrived, it was not something that was generally foreseen. 

It raises two questions.  First, how do we make our way back?  And second, what can we do to make sure that it doesn’t happen again?

As to the first, I think one has to take considerable satisfaction in the progress that has been made.  It was not unreasonable 15 months ago to fear a Depression-like pattern.  The economy was losing more than half a million jobs a month.  GDP declined as rapidly as it has in 50 years in the first quarter.  The stock market had fallen to the same level in real terms after correcting for inflation that it reached in 1966. 

We have a long way to go.  Unemployment is at unacceptable levels, and it will remain at unacceptable levels for a substantial interval.  The amount that our economy is producing is more than $1 trillion short of its potential.

But I would say to you that that sense of freefall is not our situation today. A 6 percent GDP loss in the first quarter has given rise to a 6 percent gain in GDP in the fourth quarter, albeit in part a reflection of special and unsustainable factors.  Markets, which were discounting a catastrophic outcome, have risen by more than two-thirds since April.  And there is every reason to expect that in the near future we will see a resumption of job growth.

Economists will debate, until those debating the issue become historians, at which point historians will debate, just why and how this path.  But I do not believe there is anyone who can credibly assert that it would have happened without strong government action.  Whether what was most important was the stimulus provided to spending, the guarantees that averted bank-run-type phenomena throughout the financial sector, or the substantial expansions of the central bank balance sheet can be and will be debated for a long time.

But it is, I believe, important to remember that we have been reminded by these events of a central lesson: that, for all its tremendous virtues, we cannot rely on markets to always be a self-sustaining system – that when positive feedback takes over, whether it’s the phenomenon where asset sales lead to falling prices, which leads to deleveraging, which leads to more asset sales; or whether it is the phenomenon of a weakened financial system weakening the economy, which weakens the financial system, which weakens the economy; or whether it is the reverberating doubt associated with panic that leads to a freezing-up of the financial system – there are moments when strong public action is necessary to preserve economic stability.

And the policy approach that was followed – a policy approach that emphasized fighting these vicious cycles in every way possible by stimulating spending, by forcing the raising of substantial capital, by increasing transparency, by providing a credit on substantial scale – was an effective strategy.

When I came in and was asked by President Obama to advise him in this critical period, people asked me what my objective was.  And I said that I had been struck, when my daughters took U.S. history in high school, by the fact that they took a quite comprehensive U.S. history course, but all kinds of events that seemed pretty significant to me – the 1982 recession, the 1987 stock market crash – had not been mentioned.  On the other hand, they had spent a month studying the events of the 1930s.  And I took it as my objective to make sure that this economic fluctuation was not going to be the subject of study in high school history courses a generation from now. 

And by that standard, I think there is a very good prospect for success, and certainly the prospects for success look rather better than they did a year ago.

The main topic I want to address tonight is the lessons of this crisis for longer-term economic policy and for financial regulation. 

There are some who suggest that our recent crisis calls into question the basic premises of American capitalism as an institution.

I reject this view.

Mine is closer to that of Keynes, who in 1931 famously analogized the Great Depression to a “magneto” problem.

Magneto was British English for the alternator of a car.  When it shorts out, the car does not move.  But it does not mean that the car needs to be discarded – rather, that the faulty component needs to be fixed.

Nothing that has happened calls into question the faith that properly regulated markets, freedom, and the application of science and technology together offer the most reliable route to betterment.

Equally, though, I reject the view expressed by some in the financial sector that financial crises are an inherent part of economic life that one has to simply accept and plan for.  The view that there are things that happen every five to seven years, and it’s just the way it is, just as hurricanes happen occasionally. 

It is said that John Kennedy took arms control far more seriously as a central issue after the Cuban Missile Crisis.  Equally, after the events of the last two years, we need to rethink domestically and globally a system of financial regulation that has proved manifestly inadequate.

Think about it.  During the last generation the world has seen the Latin American debt crisis, the 1987 stock market crash, the S&L debacle, the Mexican financial crisis, the Asian financial crisis, Russia/LTCM and their aftermath, the bursting of the dot-com bubble and Enron, and now the financial crisis that began in 2007.

You add it up – once every three years, a financial system that is supposed to have as its central function the distribution, management, and allocation of risk has instead proved to be a source of risk that has threatened, if not wrenched, the jobs and livelihoods of hundreds of thousands, if not millions, of people who when they hear the initials AAA think first of an automobile club and not a bond. 

Not only has our financial system failed to well-serve our economy – it has not even well-served those who have invested in our major financial companies.  An investor who invested in a diversified portfolio of major financial firms a decade ago would not only not have done better, but would have significantly underperformed a stock market that has suffered through its worst decade since the Second World War.

All of this suggests a compelling case for thinking very carefully about the future of financial regulation and in particular the problem of too big to fail.

Several observations, I think, help to frame the discussion:

  • In some areas – airline pricing, for example – there is an entirely legitimate debate not about how to regulate, but about whether to regulate.  This cannot be the focus when we’re discussing the financial sector.  Imperfect information is ubiquitous, there are large externalities to major firms’ actions, and there will always be doubts about even stating government commitments not to intervene.  There is every reason to believe that pure market solutions are not viable.
  • What should be the objective of the financial system or, more precisely, financial regulation?  Jim Tobin famously observed that it takes a heap of Harberger Triangles to fill an Okun Gap – by which he meant that the microeconomic efficiencies associated with tax distortions are likely to be small compared to the output costs associated with recessions and depressions.  It seems to me equally that, as Ken Arrow once suggested to me, it’s important to remember as we approach the financial system that marginal improvements in the allocation of capital are likely to be far less important than reducing the risks of major crises and panics.
  • It’s important also to remember that problems of misaligned incentives are not confined to the private sector.  That public as well as private actors face crucial incentive issues.  In finance, as in many areas – but I think the phenomenon is particularly pronounced in finance – there is the very profound difficulty that those who know the most about it are the least likely to be disinterested.  And there is the reality that the financial services industry this year is spending $1 million per member of Congress lobbying the Congress.  For each member of the Congress, there are four lobbyists working on financial services issues.
  • The most expensive failures in American financial history, those of the government-sponsored enterprises, can be traced to a particularly toxic combination of implicit government guarantee and excessive political involvement in rule-setting and regulation.  This suggests that as we think about regulatory approaches, we place a premium on the use of rules rather than on discretion, that we emphasize the importance of transparency, and that we use market mechanisms wherever possible.
  • Finally, I think it’s important to recognize that this is, as I already suggested, an area where there are no silver bullets.  Partly, this is because markets are by their nature unpredictable.  But more profoundly it is because proper public policy involves some inherent ambiguity.

An example perhaps suggests some of the difficulties, given on the one hand the moral hazards associated with government support, and on the other hand the dangers associated with uncontrolled panic.  Few, if any, governments are able or willing to credibly pre-commit never to pay ransom to kidnappers or to respond to threats.  Yet it would be an unwise government – and few are this unwise – that established a ransom fund and laid out in advance and in full their policy towards ransom.
In the same way, as we saw again last year, governments have tended to act when the alternative is collapse.  But reliance on government has all sorts of perverse effects: no incentive to contain risks, unfair competition between those under the umbrella and not under the umbrella of protection, competitive pressure on those who do not wish to be imprudent, who see no alternative if they are to profit. 

As the experience of Iceland demonstrates, the ultimate consequence of overly clear policies suggesting that financial institutions that are too large to fail is likely to be financial institutions that are too big to save.

What does this suggest as imperatives for policy going forward to address too big to fail?

As I look at the legislative debate in the United States and the global debate, I would suggest six imperatives.

First, comprehensive regulation of all systemically important institutions. 

Any institution that is big enough and interconnected enough that its failure can bring down the financial system is big enough and interconnected enough that it should be subject to comprehensive regulation by an accountable regulator.  Financial institutions must be regulated based on what they do, not what they call themselves. 

Anyone who has served in academic life, in business, in government has had experience with committees.  And perhaps there’s someone who will come up afterwards and explain to me why assigning accountability for getting a task done to a committee is the right way to get it done.  My life experience – and, I suggest, yours – suggests otherwise.  And that’s why assigning an institution responsibility for regulating in a comprehensive way individual institutions is crucial.

Second, resolution authority. 

The events of the fall of 2008 made clear the dilemmas that current policy saddles policymakers with.  With respect to banks – not bank holding companies, to be sure – with respect to banks – when they are on the brink of failure, we have the protocols and procedures for their resolution.  With respect to the non-bank portion of bank holding companies, with respect to financial institutions that are not banks, we have no similar procedures for managing failure.  That creates a choice between the kind of strategy that was pursued with respect to Lehman and the kind of strategy that was pursued, at great taxpayer expense, with respect to A.I.G.

Surely, if we are to avoid reliance on the expectation of subsidy, we need procedures for the management of failure of even the largest financial institutions.

This involves questions of the priority of different claims.  Since the largest institutions are almost certain to span national borders, it involves difficult issues of international coordination.  It is something that is unlikely to be done well if it is done in the heat of crisis.  And that is why one crucial piece of developing resolution authority must be insistence that institutions go through the exercise of planning for their closure in the event of crisis before crisis comes.

Our financial system will not be fail-safe until it is safe for failure. 

Third, we need to seek a regulatory regime that applies pressure to level standards up rather than to win races to the bottom.

We need to raise and make comprehensive higher capital requirements and restrictions on leverage and requirements for liquidity.  Substantial capital requirements attenuate moral hazard problems by ensuring that lenders will rely on capital rather than the perception of a guarantee.  They reduce the ability of shareholders to rely on a “heads-I-win, tails-the-taxpayers-lose” dynamic. 

And critically, properly framed liquidity requirements, attentive to systemic risk, reduce the prospect for the kind of panic we saw in 2008.

I emphasize this last point on the centrality of liquidity.  We traditionally extensively regulate the size of the reserves that institutions hold against bad outcomes.  But at least equally important is the nature of their liabilities.  An institution that has funded its activities with long-duration debt or a nation whose outstanding debt to the global economy is of long duration is much less vulnerable than one whose debt can all be demanded in a very short horizon.

Fourth, we need to recognize that there are costs associated with extensive scale and interconnection.

Taxpayers supported the extraordinary actions that were taken to stabilize the financial system, and the American people should be compensated for the assistance they provided.

That’s why the President has proposed a Financial Crisis Responsibility Fee, which would apply to the largest and most highly leveraged firms – those with more than $50 billion in consolidated assets.  In doing so, the fee would deter excessive leverage for the largest financial firms.

I must say that while I think that a great deal of the dialogue that has taken place, back and forth, between the official sector and the financial sector has been of great value in understanding these problems and in setting a course forward, I have found some of the statements coming out of the financial sector to be quite remarkable.

On successive days, it is explained that no one should have any concern about bonuses.  It’s, “Yes, bonuses are paid out, but it’s necessary in a whole set of ways, and it is to attract the right people.”  But it is inconceivable that the payment of bonuses, which after all do deplete an institution’s capital, would have any effect on its level of lending.  Then, all of a sudden, it is asserted that while payment of $5, 10, 15 million in bonuses does not deplete capital in any way that affects lending, $1 billion in taxes to pay back generous benefits that have been provided in the past suddenly represents a major depletion of capital that, since capital ratios are around 10, will reduce lending by $10 billion. 

Both propositions cannot be right.  There’s room for debate about what the right thing to do is in these areas, but I think these ideas do need to be considered on their merits.

Fifth, requiring use of clearinghouses and exchange for OTC derivatives and swap transactions where possible.

Much of the risk of runs comes from the possibility that major institutions that have provided insurance, served as a counterparty, or swapped transactions will fail.  When that risk becomes large, there is a sudden effort to close those transactions out.  The costs can be very substantial.  Moving those transactions, where possible, to clearinghouses, where a large group of institutions stands behind those transactions, can promote transparency, reduce the risk of panic, and increase stability.

Never again should we face a case like that of A.I.G., where the entire system is threatened by the actions of a single player, virtually unregulated because of the nature of its charter despite the risk of its activities.

And sixth, appropriate restrictions on activities of those who benefit from the safety net.

There are all kinds of financial activities that play a valuable role in a market economy – all kinds of speculation that confer social benefits.  No one should challenge that they are beneficial.  But not every activity that is socially beneficial needs to be carried on within institutions whose liability holders rely on the possibility of government support. 

And that’s why Paul Volcker has usefully opened up the question of restriction on the activities of banks, where those activities are not directed at serving customers, but are instead motivated by other objectives.

I’d suggest these six principles as principles that provide criteria for whether an approach, followed domestically in the United States or followed globally, is doing what can be done to reduce the risks of crisis.

I am under no illusion that we have seen the last financial crisis.  I am under no illusion that these phenomena are fully understood.  There are, for example, huge issues having to do with the relationship between stability of each institution in a financial system and stability of the system as a whole, where issues of the fallacy composition are likely to loom large.

I am convinced that the system we have today is to court excessive risk.  To build a house too close to an ocean beach that may surge is foolish.  To let there be a surge, have the house flooded out, and build another house is far more foolish.  And that’s what we will do if in wake of these events we do not act to strengthen our system of financial regulation.

Thank you very much.


Stay Connected With Email Alerts