How Geithner Can Price Troubled Bank Assets

After its chilly reception of Treasury Secretary Tim Geithner’s bank-rescue plan, the market is now worried about nationalization. As well it should. Because if the Obama administration cannot come up with a viable plan to take troubled assets off their balance sheets, major banks will not recover and nationalization–a disastrous policy–might actually become the last resort.

Many observers suggest that the Obama administration, like the Bush administration before it, is unable to solve the problem of how to price the banks’ troubled assets. If the assets are bought at their “real” values, it is said, the money the banks would get would not be enough to keep them from insolvency. But if Treasury bought these assets at the supposedly bloated values the banks are carrying them, the taxpayers would be paying too much.

In fact, neither of these statements is likely to be true. Both taxpayers and banks could come out well–and so would our economy–if the government were to buy the assets at their “net realizable value,” which is based on an assessment of their current cash flows, discounted by their expected credit losses over time. Here’s the explanation:

The accounting rules relating to assets such as mortgage-backed securities require that they be marked to market if they are held for trading, or in a category called “available for sale.” Most banks hold these assets in one of these two accounts, and so mark-to-market rules apply. What happens, then, when there is virtually no market for these assets–as has been true for at least a year? In that case, accounting rules require the banks use whatever market indicators are available.

The banks follow two steps. First, they establish the net realizable value for the portfolio. This is simply what the value of the cash flows would bring in a fully functioning market, including discounts for several factors like anticipated future losses. Paradoxically, many of the banks’ most troubled assets are flowing cash near their expected rates, and thus their net realizable values are higher than the values to which they have been written down.

The banks’ second step, after establishing a net realizable value, is marking the assets to market. And that is where the write-downs occur.

Because there are few if any buyers for these assets, their market value is much reduced. Potential buyers are not interested because there is no assurance they will be able to resell the assets when they need to. In other words, potential buyers are afraid of becoming distressed sellers themselves if their financing should disappear before the market becomes liquid again.

Thus, under mark-to-market rules, the banks must discount their assets’ net realizable values. And because of the write-downs, the banks’ apparent capital is impaired.

These facts have enormous implications for government policy. If the losses on banks’ assets are principally liquidity losses, they are temporary, and the only significant issue is whether the banks have the financing to carry the assets until liquidity returns to the asset-backed market.

And if this is indeed the case, the banks are not in any sense insolvent, and nationalization would be a huge mistake. On the other hand, if the government were to buy the assets at their net realizable values–rather than their marked-down values–this would significantly improve the capital positions of the major banks.

A hint of the true situation was contained in a remark by Vikram Pandit, the CEO of Citibank, in testimony before the House Financial Services Committee last week. He noted that Citi marks to market and that “those marks are reflected in the losses we’ve taken, as well as in our income statements and balance sheets.” But Mr. Pandit then went on to point out that the bank has a duty to shareholders: “And the duty is if it turns out [the assets] are marked so far below what our lifetime expected credit losses are”–i.e., their net realizable value–“I can’t sell [them].”

In other words, Citi has marked some assets below their net realizable value, and selling them at a price lower than that value would be unfair to its shareholders. This opens a key route to a solution for the government–buying the assets at the value that banks like Citi would be willing to sell them.

Would the taxpayer be hurt if the government buys these troubled assets at these values? Not likely. The banks have already made an assessment of the assets’ net realizable values, as Mr. Pandit suggests was done at Citi. The government can quickly verify the accuracy of these valuations, including the rates used for discounting, and can of course come up with its own lower evaluation if it disagrees.

But the important point is that if the banks’ net realizable values are even close to correct, taxpayers will not lose much, if anything, if the government bought the assets at those values. Because their cash flows determine their value, the government should be able to sell the assets in the future at roughly what it paid for them.

But the key benefit is the boost in the banks’ capital which comes from a sale now. This would eliminate doubts about banks’ solvency and free up their ability and willingness to lend again.

If this is a win-win for the banks and the government, why is it that no one has thought about doing this before? Part of the answer is that the question has been phrased incorrectly.

Most of those who recognize the problems created by mark-to-market accounting have asked that the system be suspended. This has run into opposition from the accounting industry and investor groups, who are afraid that it will be a license for banks to manipulate their financial statements.

But no change in mark-to-market accounting is necessary for the government to buy the assets at net realizable value, only a decision to buy the assets at the value they would have if there were a liquid market. Unlike a private buyer, the government does not have to worry about selling at any particular time, since it can hold the assets indefinitely.

Finally, one might ask why Mr. Geithner is claiming he needs more time to do something so simple and seemingly effective. The answer here, unfortunately, is that he seems to be barking up the wrong tree.

Mr. Geithner may believe that the banks have not written their assets down below their net realizable value. Or he may believe that others believe purchasing these troubled assets at net realizable value will be unfair to the taxpayers, and he doesn’t want the criticism if he does so. That would explain why he is enlisting the private sector to make the pricing decision.

But Mr. Geithner should check his premises. There’s a solution to his problem, the banks’ problem and the economy’s problem right in front of him.

Originally published at the Wall Street Journal and reproduced here with the author’s permission.

5 Responses to "How Geithner Can Price Troubled Bank Assets"

  1. Guest   March 6, 2009 at 5:58 pm

    Mr. Peter.Talk about barking up the wrong tree. You are like perched in International Falls. -40 degrees. You remind me of Oprah, trying to stay on whatever bandwagon gets you to the next station.Your next book deal is going to be exposed, but you might succeed and just “check your own premises.”

  2. Richard4691   March 6, 2009 at 8:26 pm

    Wouldn’t it be correct to say that the market is applying a higher discount rate to the anticipated future cash flows than the banks, reflecting uncertainty about the sustainability of the flows or about the haircuts for future losses incorporated in the bank’s net realizable value exercise? As such, the disparity might just be an information problem. What is the net realizable value of a mortgage on which a bank has initiated a foreclosure action? Well, it is the anticipated future sale price, net of all expenses of sale, discounted at the bank’s cost of money, but when will that sale occur and for how much? Well, you can get an appraisal and watch local trends. And what is the likelihood that any given mortgagee is going to default? Well, you can extrapolate from late payments and employment statistics. It would help a lot if there were a uniform set of reporting rules that servicers were required to generate monthly on every mortgage in the country. It wouldn’t be a problem to come to a uniform calculation of the net realizable value of securities held in accounts that are required to be marked to market. The problem is not with the normal market mechanisms of assigning a premium or discount to a future stream of cash flows. We might get a closer alignment of net realizable value and market prices if we have a better understanding of how net realizable value is being estimated.

  3. Guest   March 7, 2009 at 8:21 pm

    Richard4691,Your analysis seems closer to the truth than Peter’s. I am one of those who does bank ‘marking’ and I can tell you it is a moving target revisited constantly as conditions change. Anyone who can tell you they know exactly what these assets are worth is either a fool, a liar, an economist or all of the above. (I take that license since I am an economist, but I get paid for results not academic theory.) If anyone can accurately predict when the bottom will come in real estate, unemployment, etc. please let us all know so we can “correctly” price those assets. Because the risk (in everything) has increased investors are demanding a higher risk premium (capitalization rate). I do believe the tangible assets can and will be priced. That does not address all the problems however. Citi’s model fit the past moment in history but it will not work going forward. If we reinstate something like Glass-Steagall and dump the Basel capital concepts we can get back to safe banking – but without the Citi, BAC, AIG predators in our midst.

  4. Guest   March 9, 2009 at 7:21 pm

    Suspend mark to market, forget about the value of cal, nev, fl, property today. Anyone in those boom and bust states knows 5 years is a lifetime, will change hundereds of thousands of dollars over the decades. Give the banks 12 months breather and get on to the next problem, PLEASE!

  5. John Nelson   March 13, 2009 at 9:46 am

    “the government should be able to sell the assets in the future at roughly what it paid for them” — Who will buy them in the future? The reason there’s no buyer now is the market suspects these assest will see a significant increase in defaults, thus making any prediction of future value very difficult. Any prudent buyer should require a “substantial” discount to warrant the risk being taken. If the government doesn’t require that discount (risk premium), than isn’t it just another giveaway to shareholders? We saw this same example in the 90’s, when the government was criticized for sweetheart deals with no participation in the upside. I applaud Mr. Geithner’s plan which ingeniously uses the expertise of the private sector while allowing government to participate if/when higher asset values are realized in the future.