A couple of weeks ago, Yves Smith picked up on the story that the TARP Special Inspector General is investigating suspicious trades in connection with the Public-Private Investment Program. When PPIP was announced almost a year ago, there was widespread speculation about how banks and other private investors could take advantage of the program to unload toxic securities onto taxpayers (technically speaking, onto investment funds containing some private money, some public money, and a lot of non-recourse financing from the government). That story more or less faded away because PPIP never really amounted to much; banks apparently decided they were better off sitting on their toxic assets, counting on favorable accounting rules and regulatory forbearance, instead of selling them.

Here’s the relevant section from the SIG-TARP report (p. 141):

“The PPIF management company in question operates both a PPIF and one or more non-PPIF funds that invest in similar securities (i.e., mortgage-backed securities (‘MBS’)). In the case of this fund management company, the same person is the portfolio manager for both the PPIF and the non-PPIF fund. In late October, the portfolio manager directed that a particular MBS from the non-PPIF fund be sold after the security — in this case a residential MBS — had been downgraded by a rating agency. According to the company, multiple bids were received, and a quantity of the security was sold to a dealer. Within minutes of the sale, however, the same portfolio manager purchased, for the PPIF, the same amount of the same security from the dealer at a slightly higher price. Later in the day, the portfolio manager bought more of the security for the PPIF from the dealer at the original price.

“The management company involved (the identity of which is not being disclosed at this time pending SIGTARP’s investigation) asserts that there was nothing inappropriate about these trades, and Treasury has concluded that the trades did not violate PPIF rules. The facts, however, give rise to difficult questions. Was the initial purchase really arm’s length, or was the dealer aware that the portfolio manager was prepared to repurchase the securities immediately? How can a manager conclude that it is wise to sell a security at one price but then almost simultaneously repurchase the same securities at a higher price? Were these trades designed to push the risk of this downgraded security from the private, non-PPIF fund onto the taxpayer-supported PPIF?”

I wouldn’t necessarily assume wrongdoing. For one thing, the fund manager is just a fund manager, meaning he makes fees (based on assets under management and performance) from both the PPIF and the private fund. So arguably, shifting losses from one fund to another doesn’t help him that much. Of course, there are various scenarios under which it would benefit him: the fees from the private fund could be higher; one fund could have profits from which he gets a cut while the other may not; he could have his money in the private fund but not in the PPIF; he could have a nudge-nudge wink-wink agreement with the private investors; and so on.

For example, on page 508 it lays out the details of Warren Buffett’s October 2008 proposal for a “Public-Private Partnership Fund,” which would eventually become the PPIP announced by Tim Geithner in March 2009.  I knew that Buffett, Bill Gross, and Lloyd Blankfein had supported the idea, but I didn’t know the details. Buffett’s idea was slightly different from the eventual PPIP.

PPIP ended up having two flavors. In the toxic loan version, the equity would be split 50-50 between private investors and Treasury, and then the FDIC would provide leverage via a non-recourse loan (technically, I think it was some kind of loan guarantee); in the examples, it would be six to one. Buffett, by contrast, proposed leverage of four to one. Like PPIP, money would go first to pay off the government loan. However, then private investors would get all the money, until they had gotten their money back plus the same interest rate that the government got. After that point, investors would get 75% of the upside, and the government would get 25%.

I was curious about how the payoffs differed under these two proposals, so I graphed them. Note that this is for the legacy loans version of PPIP; leverage ratios do matter.

Thin lines are the Buffett proposal, thick lines are PPIP; blue is private investors, red is the government. I assume a single period and a 5% interest rate, but the interest rate doesn’t affect the shape of the curves. Private investors get a lot more upside under PPIP, but that’s basically because they have a lot more leverage (and hence get wiped out faster on the downside). In PPIP, they contribute 1/14th of the money and get half of the upside; under Buffett’s proposal, they contribute 1/5th of the money and get 3/4 of the upside. Curiously enough, the government also does better on the upside with PPIP. That doesn’t seem possible, except that the government is putting up a larger proportion of the money in PPIP than in Buffett. So moving from Buffett to PPIP, the returns for private investors and the government both go up, but that’s because the weighting is shifting from private investors (the higher returns in either case) to the government (the lower returns).

In the toxic securities version of PPIP, the leverage ratio was lower, bringing the thick blue line down closer to the thin blue line.

There’s nothing too scandalous here that I can see. But I thought someone else who made it to page 508 of Too Big to Fail might have had the same question, so here’s the answer.

By James Kwak

The U.S. government does few thing better than create debt.  After a year of talking about it, the government is going to have the chance to throw their good debt, Treasury bills notes and bonds, after bad, non-performing toxic loans and securities.  The Federal Deposit Insurance Corporation (FDIC) and the U.S. Treasury are going their separate ways on their cash for trash schemes at this point.  Accountants and investors should be wary of the big prices they see coming from the FDIC’s auctions, but taxpayers should be afraid of the U.S. Treasury’s efforts to re-inflate the securitization bubble.

The FDIC is nearing the century mark of bank failures this year and it has a lot of bad assets to unload.  On Tuesday, October 6, 2009, it announced that it sold a $4.5 billion, festering pool of condo loans from the failed lender Corus Bank.  The last 10-Q for the failed Chicago lender said that it had $3.3 billion in non-performing loans with heavy concentrations in busted condo markets of Miami and Los Angeles.  Yet, zero-coupon, FDIC-guaranteed debt and a billion dollar line of credit led to the price of $2.8 billion.  That price of about 60 percent of par seems rich when almost 70 percent of the loans are non-performing.  A few weeks earlier the FDIC did its first Legacy Loans Program auction.  My paper “Slicing the Toxic Pizza: An Analysis of FDIC’s Legacy Loan Program for Receivership Assets” found that the nearly 6-to-1 government subsidized leverage in the first Legacy Loans Program sale boosted prices by over 20 percent.  Yet, the irony is that these subsidies don’t help the FDIC and ultimately taxpayers because the FDIC is offering cheap financing to sell assets it already owns.  Any higher prices just offset the subsidized financing if the auctions are competitive.  Yet, any marks obtained from these auctions are certainly inflated.

The Legacy Securities Program run by the U.S. Treasury with an initial taxpayer outlay of $30 billion is set to launch soon.  More asset managers are closing their investment funds every week.  The problem with the Legacy Securities Program is that the government will probably use cheap leverage to finance the sale of trash assets that it does not already own.  That means that the taxpayer subsidies are enjoyed by the banks selling the assets.  My research shows that the most troubled banks will be reluctant to part with their toxic securities.  Yet, the healthy banks will be all too eager to unload those assets at inflated prices.  Come Halloween the U.S. Treasury will be handing out the goodies.  Unfortunately, taxpayers will be getting none of the treats but all of the cavities.

By Linus Wilson

From the Times article: “[FDIC] officials announced that they had reached a deal to sell $1.3 billion in mortgages from Franklin Bank, a Houston-based lender that failed last November and was taken over by the F.D.I.C.”

Konczal, of course, also caught this:

“The first argument is that it would take banks that were otherwise healthy and allow them to start lending again in the middle of a credit crunch. We taxpayers pay to help unload these loans onto other private markets, so that the bank can start lending again.

“But as financeguy points out, the bank in question, Franklin Bank, is dead. FDIC took it over around a year ago, with its balance sheet deep in the red and after losing double digits in loans since 2007. It took a huge gamble in the real estate market, and it got destroyed. This isn’t an otherwise healthy bank with one bad asset on it.”

But I think he is altogether too cool about this. Let me try for a little more indignation. Our government is providing large subsidies to private investors to buy toxic assets. The only possible justification for these subsidies is that they are necessary to restore health to the banking system, by taking toxic assets off the balance sheets of banks. But these toxic assets are already the property of the U.S. government. This means that the government owns 100% of the upside and 100% of the downside on those assets.

Or at least it did until last week. Then it gave half the upside to an investment fund – “Residential Credit Solutions of Fort Worth, a three-year-old company founded by Dennis Stowe, a veteran of the subprime mortgage industry” – and kept all of the downside to itself. What could they possibly have been thinking?

(And wasn’t there supposed to be an auction in there somewhere? The Times story doesn’t mention one (although that doesn’t mean there wasn’t one). Without an auction, how do we know this isn’t more of a giveaway than it already looks like?)

Update: Thanks to commenter On the Mountain, who points out that there was an auction, and provides a link to the FDIC press release. So that part of the deal is in the clear.

By James Kwak

The Federal Deposit Insurance Corporation indefinitely postponed a central element of the Obama administration’s bank rescue plan on Wednesday, acknowledging that it could not persuade enough banks to sell off their bad assets. . . .

Many banks have refused to sell their loans, in part because doing so would force them to mark down the value of those loans and book big losses. Even though the government was prepared to prop up prices by offering cheap financing to investors, the prices that banks were demanding have remained far higher than the prices that investors were willing to pay.

I don’t think I’ve ever done this before, but . . . Simon and I, March 24:

The problem in the market today is that the prices demanded by the banks are much higher than the prices that private buyers (hedge funds, private equity firms, sovereign wealth funds) are willing to pay. The government has no way to bring down the banks’ minimum sale prices . . .

The subsidy may not be sweet enough to close the deal. According to one analysis, a specific mortgage-backed security was held on a bank’s books at 97 cents, while its market price was about 38 cents. Even if you limit the buyer’s potential loss to the capital he put in, it’s unlikely he will raise his bid from 38 cents to anything near 97 cents. . . .

Just last week at least some banks wanted to participate in the program – to buy assets from themselves. Once Sheila Bair rejected that idea, I guess they lost interest. Essentially the stress tests placed a big government stamp of approval on their balance sheets, so their current strategy is to wait out the recession and hope the prices of their legacy loans recover. There’s no downside risk, because if the economy gets worse and they ever need to unload those loans, they can count on the plan being resurrected.

I guess we were, however, wrong to worry about inter-bank collusion in the legacy loans program.

(Note that this does not apply to the legacy securities program, which may still be going ahead.)

By James Kwak

Although banks cannot buy their own assets, Bair did say, “I think there have been separate issues about whether banks can be buyers on other bank assets and I think that’s an issue that we continue to look at.” As I said yesterday, and as Winkler also said, I think this is also a bad idea. Even if you successfully deter outright collusion, you can still have outcomes where the industry as a whole is using subsidies to overpay for its own assets and shift the loss onto the government.

And no, I don’t actually think that Sheila Bair reads this blog, much less listens to what I have to say.

By James Kwak

To recap: The Public-Private Investment Program provides subsidies to private investors to encourage them to buy legacy loans from banks. The goal is to encourage buyers to bid more than they are currently willing to pay, and hopefully close the gap with the prices at which the banks are willing to sell.

Allowing banks to buy their own assets under the PPIP is a terrible idea. In short, it allows a bank to sell half of its toxic loans to Treasury – at a price set by the bank. I’ll take this in steps.

Assume that a bank has a portfolio of loans on its books at $100 and that, when held to maturity, those loans will turn out to be worth $90. In the absence of any transaction, it will end up losing $10.

First, imagine the bank is buying these loans from itself with no government support. Let’s say it decides to pay itself $80. Obviously this does nothing – no cash has changed hands, and the bank still has the toxic loans. It might take a $20 “loss” today, but it will gain $10 by holding those loans to maturity, so its net loss is still $10.

Second, imagine that the bank creates an investment fund where its own money is matched dollar-for-dollar by Treasury money. If the fund buys the loans from the bank at $100, then the fund will eventually lose $10. That loss will be split between the bank and Treasury, so now the bank’s net loss is only $5, because the other $5 of losses is absorbed by Treasury. Conversely, if the fund pays less than $90 for the loans, the gains will be split with Treasury, so the bank’s net loss will exceed $10. If it pays exactly $90, then the bank is no better or worse off than before the transaction.

In other words, it’s exactly the same as if half of the loans were sold to Treasury, with the bank holding onto the other half. From the bank’s perspective, it wants the price to be as high as possible.

Third, imagine that the equity in the investment fund is split evenly between the bank and Treasury, but it is leveraged up to six-to-one by a loan that has an FDIC guarantee. This has the effect of capping the bank’s downside if the loans do really badly. The guarantee wouldn’t kick in if the loans end up being worth $90, but let’s say they end up being worth $70. Now if the fund pays $90 for the loans, that is about $6.50 of bank money, $6.50 of Treasury money, and $77 of FDIC-guaranteed loan. So the bank loses $6.50, Treasury loses $6.50, and the FDIC loses $17. With no transaction, the bank would have lost $30. 

The third scenario is the PPIP.   

So if banks are allowed to buy their own loans, they will have the incentive to overbid for those loans. As long as the price they pay exceeds the eventual value of the loans, they will come out ahead, even without the loan guarantees. This is bound to close the gap between buyers’ bids and sellers’ reservation prices – by ensuring that at least one buyer (the bank) will bid more than the reservation price set by the seller (the bank). And the way these investment funds are set up, the private equity partner – the bank – will be the one deciding how much to bid; the whole point of these “partnerships” was to get the government out of the business of valuing assets. In short, since the government doesn’t have the expertise necessary to guard the henhouse, we’ll let the fox do it.

Do you think this is so crazy it couldn’t possibly be what the banking lobby is asking for? Consider this:

[Norman R.] Nelson proposed to the FDIC that banks be allowed to control as much as half the capital in a buyers’ group. In some cases, he wrote, “the selling bank should be able to participate as the only private-sector equity investor.”

(Nelson is general counsel of the Clearing House Association, a trade group representing ten of the world’s largest banks.)

The blather being spouted in support of this self-dealing is so blatantly disingenuous it makes you wonder if this is some sort of joke.

“Banks may be more willing to accept a lower initial price if they and their shareholders have a meaningful opportunity to share in the upside,” Norman R. Nelson . . . wrote in a letter to the FDIC last month.

Um, no. Participating as buyers will only ensure that banks will be motivated to overpay. If they think that the price is too low, they would prefer to just hold onto the asset – especially now that the stress tests have made clear that no bank will be forced to liquidate assets under pressure.

“Bankers see it as a win-win,” said Tanya Wheeless, chief executive of the Arizona Bankers Association, which has urged the FDIC to let banks buy their own assets through PPIP.

That’s absolutely true – for them, that is.

Towne Bank of Arizona plans to sell some of its soured real-estate loans into PPIP and wants to profit from the program. “We think it would be attractive to our shareholders to be able to share in whatever profits there are from the venture,” said CEO Patrick Patrick.

It seems like this proposal is too extreme for some people in the banking lobby to support; neither Scott Talbott of the Financial Services Roundtable nor Edward Yingling of the American Bankers Association was quoted.

You would think that the government would slap down this idea in a second. But here’s the FDIC spokesman on the issue: ”It’s an issue that’s been raised and an issue we’re aware will need specific guidelines.” 

Now, even if banks aren’t allowed to buy assets directly from themselves, there are still reasons to be worried. If you think of the banking sector as one big entity, then it’s obviously in the interests of that entity to buy assets from itself exactly as described above – only with Bank A buying from Bank B and Bank B buying from Bank A (or more complicated permutations as required). In other words, there is a massive incentive to collude. Now, collusion is illegal. So the question is whether the banks can get themselves into an equilibrium where they all overpay for each other’s assets – thereby benefiting everyone – without actually conspiring to do so. This is like when one airline raises prices and immediately every other airline follows suit – there’s nothing illegal about it, even though the end result is the same as if they had colluded. 

But let’s not make it easy for them. If this proposal has any chance of going anywhere, then Tim Geithner or Sheila Bair should come out and reject it right now.

By James Kwak

Snower’s hypothetical asset has an expected value of $50. According to his calculations:

• If the bank has it on its books at $70, the private sector will bid it up to $85 because of the government subsidy. The government would have been better off under the original Paulson Plan (just buy it off the bank at book value, in this case $70).
• If the bank has it on its books above $85, the private sector will not buy it at all and the plan will do nothing.

Now, his asset has different characteristics than the assets out there in the real world, whose expected values are not knowable, let alone known. That may change the analysis, but I doubt it changes the ultimate result.

Thanks to the reader who recommended this.

By James Kwak

The problem is that bankruptcy is the only weapon Chrysler and Treasury have in this fight, and it’s a strategic nuclear weapon. Bankruptcy is the only threat that can get the bondholders to agree to a swap; but because a bankruptcy carries some risk of destroying Chrysler (because control will lie in the hands of a bankruptcy judge – not Chrysler, Treasury, the UAW, or Fiat), and taking hundreds of thousands of jobs with it, everyone knows that Treasury would prefer not to use it. The bondholders are betting that they can use Treasury’s fear of a bankruptcy to extract better terms at the last minute. (And it’s even possible that the large banks agreed to the swap knowing they could count on the smaller, less politically exposed hedge funds to veto it.) But Treasury may still press the button, because it needs to make a statement in advance of the bigger GM confrontation scheduled for a month from now.

But there’s a much bigger, slower game going on at the same time, and the administration’s basic problem is the same: all it has is strategic nuclear weapons that it absolutely does not want to use. The New York Times had an article today about how “a growing number of banks are resisting the Obama administration’s proposals for fixing the financial system.”  It didn’t have a lot of new information, but it summarized the outlines of the game.

The administration has created three main tools to help the banks – and it really, genuinely wants to help them:

1. The Capital Assistance Program (CAP) to give new capital to banks that need it (based on the stress tests).
2. The Public-Private Investment Program (PPIP) to encourage the private sector to buy troubled assets from banks.
3. The Term Asset-Backed Securities Loan Facility (TALF) to provide funding to the asset-backed securities market, which is being expanded to mortgage-backed securities.

According to the Times article, the banks want none of it – even though, as many people (including us) have argued, the terms of these programs are clearly favorable to the banks.

Instead, banks such as Bank of America and Citigroup are arguing that they are more sound than the stress tests indicate. This claim is almost not worth debunking, but I’ll give it a few words anyway. First, the “more adverse” macroeconomic scenario used in the stress tests is already more optimistic than the forecasts of respected bodies, such as the OECD. Second, the IMF recently estimated total losses on financial institution balance sheets at $4.1 trillion, future writedowns by U.S. banks at $550 billion, and U.S. bank capital needs at $275-500 billion. If B of A and Citi don’t need the capital, who does?

In addition, according to the Times, “Several big banks have declared they have no intention of participating in the [PPIP]. . . . Many banks are reluctant to sell their nonperforming loans because they could suffer big losses, forcing them to raise more capital.” Finally, only $6.4 billion in loans have been given out under the TALF – a program currently sized at $200 billion, and projected to grow to $1 trillion.

Why are the banks turning their banks on this government largesse? I think there are two reasons.

First, taking capital under the CAP or selling assets into the PPIP involves some hardship, despite the taxpayer subsidies involved. Raising capital dilutes existing shareholders, and selling assets (at prices where someone will buy them) will require writedowns from their current, unrealistic book values. Treasury really wants the banks to participate, because it will increase confidence in the banks, and that’s why Treasury is offering to share the pain, via underpriced capital and low-risk loans.

But even though Treasury is so generously offering to share the pain, what’s the incentive for the banks to suffer any pain at all? We know the government won’t use the strategic nuclear weapon and let them go bankrupt or pull their banking licenses (which amount to the same thing). And Tim Geithner’s request for a battlefield nuclear weapon – resolution authority for systemically important financial institutions, including bank holding companies – seems to be going nowhere in Congress. This is not surprising, since the banks have already demonstrated that they can count on most or all Republicans and at least a few Democrats in the Senate. With the administration’s hands tied and the banks’ political power intact, the banks are in the same position they always were: if things go well, they will make money; if things go badly, the government will always bail them out later, on terms they are willing to accept.

On the one hand, the banks are complaining about unprecedented government interference and pressure, and to some extent that is happening. But on the other hand, the banks are ultimately calling the shots, because they know Tim Geithner can’t use his only real weapon.

Second, the incentives of managers and shareholders are not aligned. A major factor in the banks’ reluctance to participate in their own rescue seems to be fear of government interference, which is code for executive compensation restrictions. 

Executives worry that whatever assurances the White House gives them, an angry Congress might impose new rules on banks that participate, particularly on pay. . . . “We’re certainly not going to borrow from the federal government, because we’ve learned our lesson about that,” [Jamie Dimon] said earlier this month in a conference about earnings.

Now, while I think some of the compensation caps discussed in Congress (but not passed by the Senate, as far as I know) were silly, I haven’t heard a lot of shareholders complaining about them; it’s the managers who don’t want them. So the situation is very simple. Participating in PPIP, for example, might be a net positive for shareholders, because even though it forces short-term writedowns, it also reduces the risk of larger writedowns in the future. But if managers think that it will lead to compensation limits, then it is a net negative for managers. I think our readers can fill in the rest of this thought.

By James Kwak