Thursday, December 9, 2010

More on TARP

Karl Smith has kindly responded to my post on TARP:

Then a broker-dealer went bankrupt. The process of chaos began to unfold. We had what looked very much like a run. Given that narrative the response was to somehow stem the run. This meant insuring creditors against losses.

Various schemes were proposed. In the end equity injections were settled upon. The terms where not what I would have wanted but I understand the pressure of being in the moment.

After the equity injections our measures of panic began to subside and indeed have not come back despite the fact that housing has not recovered. This lends credibility to those of us who said that housing and bad loans mattered only to the extent that they were possibly creating insolvency and fueling a run.

Do we know for sure that this is narrative is correct – no. However, the level of certainty I think Gupta is asking for is virtually never available to us in real time and is hard to ascertain even looking back. My core case is that the narrative holds and continues to hold as we accumulate more data.
The ultimate argument here is "Things were going very bad; we did something; and things ended up much better than we expected. Maybe things could have gone better, but things did end up going surprisingly well."

First -- there's the question of whether or not TARP "helped." Ultimately, this relies on the unknowable counterfactual of what the world looked like without TARP. It also relies on assuming that the entire recovery was due to TARP. Yet as Donald Marron points out, "One reason that TARP appears much less expensive than originally predicted is that many of its investments benefitted from other government actions whose costs show up elsewhere in the budget." We did a whole lot to help the banking sector, and would undoubtedly have done even more without TARP. So to give it the whole credit is probably a bit much.

There's also a key phrase here -- "in the end equity injections were settled on." This is an interesting way of putting it. What actually happened is that Treasury asked for a virtual blank check ostensibly to support asset purchases, and then decided to purchase bank equity instead because they wanted to.

It's worth emphasizing again that the narrative that Treasury was selling -- that the world would collapse if we didn't purchase toxic mortgage assets -- was completely wrong. In fact we didn't purchase many toxic assets, and the world went on. Had we enacted TARP as envisioned, written into law, and sold to the public -- the result would have been hundreds of billions of dollars of taxpayer losses.

So the fact that we "settled on" equity injections reflects my primary opposition to the bill -- that we basically left the decision of how to spend $700 billion dollars in the discretionary hands of the Treasury. We are indeed fortunate that Congress failed to implement any meaningful oversight on the bill and the ultimate decision Treasury opted for was not the worst thing in the world (though, of course, TARP money has also been used as a political slush fund for things like HAMP, auto bailouts, etc.). But what was at stake in the bill was really the democratic principle that the legislative branch, not unelected bureaucrats, decide what we spend money on. Do we really think that giving Treasury the right to spend hundreds of billions of dollars is a good idea? Can we continue to rely on them reading the right bloggers?

I am sympathetic to this view though I take a slightly different tact. I tend to think that the government should have simply squeezed the financial system for everything that it was worth on the grounds that its the responsibility of the Treasury Secretary to act in the best interest of his clients. The taxpayers were his clients. Making more money for you clients is generally preferable to making less thus he should have tried to make more money.

However, it is important to note that TARP as structured was profitable on the bank side. It wasn’t simply that the taxpayer got his or her money back. Their were warrants that that gave the taxpayers a bit of the upside. I just don’t think they were big enough.

I understand the moral outrage and all but my general take is that the Government in this capacity should act to maximize taxpayer profits, not express taxpayer outrage. I realize that this is not a majority view.

Since when is the government supposed to run a hedge fund? Since when should the maximization of "taxpayer profit" be an objective of government?

The issue here is why is the government doing this at all? If the government made money on at least some of its investments; and the banks did well; why couldn't private capital flow in?

Everyone agrees that banks in 2008 needed to be recapitalized. But it was an open question whether this would be achieved through bankruptcy, or some mechanism of forcing banks to raise capital, or so on. There were a lot of options, each with costs and benefits. Some of them wouldn't have involved public money.

Unfortunately, it looks from Swagel's account that Treasury -- though aware of this problem -- spent months doing nothing about this. They wound up in September with a three-page plan that, as I've mentioned, would have barely helped. Perhaps they had the excuse that they were rushed and didn't have the time to properly consider, say, Zingales' bankruptcy proposal. But it's been several years since then, and Treasury doesn't have any serious bank resolution mechanisms that don't require public money.

Sunday, December 5, 2010

TARP was no Success

As over two years have passed since the extraordinary events of the worst days of the financial crisis in 2008, a new wave of retrospective and judgments are coming out to evaluate the nature of the federal government response.

In particular, a number of commentators have remarked favorably on the TARP program. At the time, widespread financial collapse seemed a distinct possibility. The immediate injection of large amounts of public funds, though unpopular, was followed by a relative stabilization of financial markets. The money spent on the program may even be paid back in large amounts.
Yet there are several reasons to be skeptical of TARP-revisionism:

1. No Silver Bullet. The case that TARP was a successful program of equity injection is based on praise by association. TARP was passed; the financial sector seemed to revive itself; therefore TARP must have fixed the financial sector.

Yet it is impossible to causally trace the improvement of conditions in the financial sector to any one program. The federal government also implemented a number of other programs to ease conditions for financial firms — from increasing access to the Fed’s discount window (resulting in trillions of dollars of loans to insolvent institutions), to easing mark-to-market accounting rules that allowed banks to hide losses, to unprecedentedly low interest rates which allowed banks to accept cash deposits from customers while paying virtually nothing. The scale and scope of the federal government’s interventions in the banking sector were enormous, and TARP does not deserve the entire credit for turning things around.

And though the situation remains better than in the worst moments of the financial crisis, the supply of credit remains weak. Small businesses report difficulties in obtaining credit. The mortgage lending market is almost entirely handled by the government; while the securitization market — a large financing source during boom years for everything from mortgages to student loans and airplanes — remains a shadow of its former self.

2. Costlier than you think. Suppose Hank Paul went to Las Vegas and put down $700 billion of the government’s money on the roulette table. Suppose he happened to win; though the casino only allowed him to take back the amount of money he brought with him. Would we think this provided an acceptable rate of return?

The answer is obviously no. TARP was wagered on a “heads-I-win; tails-you-lose” basis. If banks had done worse than they actually did, the taxpayer would be on the hook for hundreds of billions of dollars. If the banks did well (as they subsequently seem to have); the government merely gets it money back.

In other words — the government was not provided an adequate risk-compensated return. In backing the American financial system, the Treasury Department took on an enormous financial gamble on behalf of the American taxpayer, one that could easily have gone bad. It is fortunate that things did, in fact, go well. But that doesn’t prove that the original risky gamble was sound; only that taxpayers were lucky, and under-compensated for their investment.

Nor is it even obvious that the government will, in fact, actually recoup its entire investment. Money lent to GM and AIG may never be repaid in full. Loans to a number of financial firms were repaid in the form of equity, which may not recover in value. And, of course, the faster that firms pay back TARP money, the less effective the program will be in actually improving bank balance sheets. Finally – there are the long-term costs. Now that the government has established a precedent for bailing out firms that are Too-Big-to Fail (which include domestic automobile companies), similar companies will expect comparable guarantees in future crises. This will encourage risky lending, increase the probability of future crises, and lead to further taxpayer-fueled bailouts in the future.

3. Horribly Conceived. Most of the discussions of TARP, including this article, focus on the manner in TARP was executed: in the form of equity injections in financial firms and debt guarantees. It is easy to forget that, initially; TARP was conceived in a radically different manner.

As its name (Troubled Asset Relief Program) suggests, the program was designed in order to purchase toxic assets, in particular poorly performing mortgage-backed securities. The idea was that a handful of bad assets were “clogging” the financial system, and were really worth more than market price. By purchasing these assets at above market value, the government could assist financial firms while creating a viable market for these assets.

In retrospect, this idea was clearly and disastrously wrong. Mortgage-backed securities were not cheaply priced because investors were “panicking”; rather, it was because they were worthless. If anything, they were priced too high in view of the subsequent collapse of the housing market. If TARP were executed as planned and authorized by law, taxpayers would have lost hundreds of billions of dollars purchasing toxic assets.

Of course, it is easy to say so in retrospect. Yet even at the time of the program’s announcement, a large group of eminent economists wrote a letter to counter the program. Many of these figures were banking experts who were not averse to government support of the financial system. But they were all opposed to the idea of handling the government’s role through the direct purchase of bank assets.

4. Subversion of the Democratic Process. If both the public and economic policymakers were opposed to the program, how on earth did TARP pass? How was it transformed into a program of equity injections? This is real problem with TARP: that it represents a form of undemocratic discretionary bureaucratic overstepping.

In late September two years ago, Treasury Secretary Hank Paulson went to Congress. His message was essentially: “The financial system is going to collapse. We need $700 billion dollars. Trust us.” Unsurprisingly, further financial crisis ensued. Yet the following crisis merely increased the feeling of crisis and panic, further increasing the hand of the Treasury Secretary.

Fortunately, over time, the asset purchase aspect of the program was shunted to the PPIP, and then dropped entirely.

Yet this only means that money budgeted for one purpose was directed to another. While Treasury was authorized to spend TARP money as it saw fit; it ultimately dispensed funds in a different manner than advertised. Rather than going to Congress to meet budgetary needs for a fixed plan, Treasury viewed the original $700 billion as a grant to spend as preferred. There is some evidence that TARP funding went out to banks in proportion to their political ties. This type of crony capitalist insider trading is not conducive to a proper financial recovery, or public trust in the financial system.

The redirection of TARP funds did not stop there. GM and Chrysler qualified for relief under TARP, though both are automakers rather than financial firms. Over the past two years, TARP has been used as a general slush fund to handle any sort of general government spending that proved inconvenient to finance democratically through the regular budgeting process. For instance, HAMP was created using $40-50 billion dollars of TARP funds, yet has proved to be a dismal failure. Few Americans will stay in their homes because of HAMP modifications; most will only experience a delay in the foreclosure processes. Granting this sort of broad mandate and extraordinary funds to unelected government bureaucrats is, as the humanities types say, “problematic.”

Incidentally, the fact that Treasury was able to change its favored strategy after a matter of weeks and drop the idea of large-scale asset purchases altogether suggests that their preferred narrative of the crisis – that outside events pushed them into certain catastrophe-prevention measures – cannot be the case. Surely if the world was about to fall apart without large-scale government intervention, yet nothing like that happened in the absence of government intervention; the folks who saw government intervention as necessary were less than perceptive. Surely if the Treasury were in fact acting under the presence of hard constraints, it would not be able to change its favored policy completely in a matter of weeks.

5. There were Better Alternatives. Few banking experts would doubt that recapitalizing the banking sector is an essential priority in the aftermath of a financial crisis. When banks do not have capital, they cease lending, and the economic recovery remains tepid.

Yet at the same time that TARP was proposed, better alternatives were floated. Garret Jones, an Economist from GMU, proposed the idea of a “Speed Bankruptcy” that would recapitalize financial firms by converting debt into equity in a forced swap under a bankruptcy-like procedure. This would allow for banks to be bailed out by private capital, rather than public money. It would have forced bondholders — who largely went through the crisis unscathed — to bear the risk for which they enjoyed the return. Luigi Zingales advocated for similar proposals – and Phillip Swagel’s account of his time in the Treasury makes it clear that these ideas were seriously considered. There were plenty of ways to induce a private sector recapitalization of the banking system, from reforming the bankruptcy code to simply ordering banks to accumulate more capital by fiat. These steps may have involved going to Congress or skirting the law. They may have involved a different set of cost-benefit calculations. Yet as Luigi Zingales notes, nowhere in Swagel’s account do we see evidence that Treasury properly weighed the costs and benefits of various bank recovery options in a systematic manner. Everything instead hinges on various political “constraints” that apparently make an amendment to the bankruptcy code impossible, but permit Treasury to demand a $700 billion blank check from taxpayers.

In fact, Treasury was hardly the white knight of the crisis, working with distinction to do the (presumably unique) right thing in the face of the idiots in Congress. They actively shaped the narrative of what was going on; presented only the solutions they favored; and only saw legal obstacles as binding when they applied to policies they did not favor. They may well have been patriotic workaholics as well. That only suggests that no such agency, well-staffed though it may be, should assume that level of power or influence.

Ultimately, the difficulty in evaluating a program like TARP lies in defining the proper benchmark. Many commentators have mentioned the government recouping its investment as a sign of the program’s success. Yet as Felix Salmon points out; earning a financial return is neither a necessary nor sufficient condition of a successful government intervention. A financial rescue operation that “cost” the government money may have been successful if it were paired with substantial mortgage modifications, recognition of bank losses, and an eventual recovery in lending. A rescue operation that earned the government a substantial pile of revenue may prove to be a failure if money were simply directed at banks that were already financially healthy.

In fact, there may be a contradiction between the goals of recouping government investments and assisting financial sector recovery. Investments in healthy banks – like Goldman Sachs – may maximize the government return; yet do little to foster overall economic recovery. On the other hand, lending to risky banks – like Citibank – may prove more financially risky. Yet exactly because Citibank faces tangible economic risks, lending to it provides the chance to shore up a struggling institution and raise the odds of a financial recovery.

The ultimate lesson from TARP is that rushed responses to severe events rarely pan out as their architects intended. Building a financial system that never breaks down is impossible; yet building one that fails gracefully is in our power, if we embrace rule-based and punitive bank resolution techniques like bail-ins, instead of discretion-centered responses that demand public money and fuel moral hazard.

Wednesday, December 1, 2010

Sentence of the Day

Germany, like China, is less prosperous than it seems, because its surplus production is geared to sale for claims that cannot credibly be redeemed for what the country’s citizens would want should they exercise their option to consume.