A Geithner Plan for Europe

 The lending performance of Irish and other European banks might be improved by the creation of asset management companies to absorb banks’ toxic assets and replace them with cash or near-cash assets.  The types of toxic assets held by banks differ across European countries; they are mostly bad property loans and collateralized-mortgage-based securities.   

There will be a considerable decrease in banks’ accounting book value when it sells toxic assets for cash value since the toxic assets are being carried on the banks’ accounting books for more than their true market value.  The decrease in accounting book value, which will come out of the banks’ book value of equity, has to be modest enough so that the banks are not declared insolvent after the transaction.

The Geithner plan offers a valuable template for Europe in designing a toxic asset transfer scheme. The Geithner plan invites private firms to bid competitively for the toxic assets of US banks. The funding for the bids come from three sources: equity provided by the bidding firm of 3% to 10% of total capital provided, an equal amount of equity funding provided and owned by the US government, and a non-recourse US government loan for the remaining 80-94% of capital provided.  The required yield on the nonrecourse loan will be somewhat underpriced  relative to its risk, so there is some degree of subsidy.  This is necessary to make the plan work; this entire subsidy should in theory accrue to the selling bank and shore up its capital base.  This is the only subsidy in the plan.

Since the US government will own up to 97% of the asset management companies’ assets, there is considerable financial/administrative/legal expertise needed by the government to provide reasonable oversight of these asset transfers and subsequent management of the assets.  This could make such a plan problematic if implemented by Ireland on its own.  Also, it is critically important that there are multiple bidders, competing aggressively against each other in the toxic asset auctions.   Additionally, the nonrecourse loan must come from a government entity borrowing at risk-free rates (and then lending to the asset management company at a higher rate).  It is a clever type of “targeted quantitative easing” (my own term).  Ireland is not in a position to borrow vast sums to purchase the toxic assets of its own banks.  The targeted quantitative easing should come from the sovereign issuer of our currency.

This is a great opportunity for the European Central Bank to play a role as a true regional central bank.  It can easily provide the non-recourse debt in a European version of the Geithner plan.  National governments can be responsible for the matching equity investment, and the ECB can provide the oversight, along with national governments.  Implementing the Geithner plan in Europe would also appease the US government which wants to see some burden sharing by European governments in dealing with the credit-liquidity crisis.

Krugman feels that the Geithner plan, although moving in the right direction, is too weak to be effective. However if the EU joined the US and instituted a very similar plan there might be a positive-feedback effect on confidence and liquidity which could boost the global impact.     

16 replies on “A Geithner Plan for Europe”

Let’s imagine a banking system in which assets are €10 billion less than liabilities, i.e. they are insolvent. Here are two plans that fix the problem and produce a banking system with positive equity capital of €15 billion:

1. Admit the banks are insolvent, nationalise them rather than shut them down to minimise disruption, admit the true value of the assets and pay off the net debt of €10 billion, put in $15 billion of taxpayer money to create cleaned-up well-capitalised banks. Sell these banks to private investors to recoup the €15 billion. Cost to the state = €10 billion. Puts the banks in new ownership.

2. Get the government to overypay for assets to the tune of €25 billion. Cost to the state = €25 billion. Existing ownership continues to control banks.

Both plans “work” but they have other implications. The Geithner plan that Greg strongly endorses is a complicated version of Plan 2 (Bacon’s proposals for Ireland appear to be a more transparent version). Personally, I prefer Plan 1.

Karl: Plan #1 does not work as well as the Geithner plan (plan 2 I guess) since the banks are still encumbered by the toxic assets and will behave accordingly, avoiding new lending to preserve capital. Better to warehouse the bad loans outside the working banking system, which we want to lend enthusiastically to new customers.

I agree that it is a shame that the plan cannot induce more changes in senior management and boards at the banks. Perhaps that can be viewed as a separate problem. Something that should continue to be pursued — perhaps the plan can have some oversight provisions regarding corporate governance and senior manager compensation.

In terms of bank “ownership” and distributional justice that is complicated by the substantial turnover of banks’ shares by institutions and even individuals. The “owners” who have held shares for more than one year (say to early 2008) are pretty much wiped out in terms of share value and percentage loss on their investment. Anyone who buys shares now should benefit more if a Geithner-type plan is adopted rather than plan #1 of full nationalization. Current owners (having held for one year or so) are sitting on losses of up to -90% and will still have big losses. Only very very long-term “owners” have a positive return over their holding period. A lot of individual owners I suspect are wealthy old people, plus of course institutions holding on behalf of pensions, endowments and insurance companies who have the majority of these near-worthless shares. I do not think that the distributional aspect is that overwhelming; more important to worry about economic efficiency in designing this plan.

Great debate. It might be worthwhile to consider a third option for Karl’s list:

3. Get the government to overpay for assets to the tune of €10 billion. Have the government also inject €15 billion in preference shares. Net cost to the state = €10 billion (assuming preference share are priced to avoid an additional subsidy). Existing ownership continues to control banks. Private owners raise equity capital over time to retire the preference shares.

This plan avoids (temporary) public ownership. Karl makes it sound as if (temporary) political control would be cost free under his Option 1. I have doubts.

An obvious problem with Option 3 is that the market value of equity capital is likely to continue to be quite low. (I assume the effective put option of limited liability makes the market value positive before the intervention. The value of this put is increasd by the asset purchases, so there is some subsidy to existing owners. But the market value of equity is likely to continue to be low.) With little equity capital at stake, there is a risk the owners will “gamble for redemption.”

Thus there are incentive issues associated with both Options 1 and 3. The broader point is that we should carefully consider all the options before going the nationalisation route. (With even Colm McCarthy supporting nationalisation, I fear the dam has well and truly broken.)

Gregory implies that Krugman is lukewarm on this latest Geithner plan. However, Krugman is in fact strongly against it. It’s the Obama fiscal stimulus that Krugman would describe as a move in the right direction. Geithner’s plan is a step in the wrong direction.

There are many more than three options. It’s easy to create better-capitalized banks with new management and with more reliable balance sheets. Buiter is just one of many who are stating the obvious least-worst-option solution

This Geithner plan is pretty insane. It’s a scatter gun approach to give free money to already rich people with no strings attached. In short, the taxpayer should take the good, easy-to-value assets and liabilities from the likes of Anglo, along with the branch network. Such a method would create a viable good bank without a cent of state money. Read Buiter’s piece for more details on this and why it’s no brainer step in the right direction.

Why buy the crud assets, when it’s the sound parts of the business that we actually care about?

Greg argues that “Plan #1 does not work as well as the Geithner plan (plan 2 I guess) since the banks are still encumbered by the toxic assets and will behave accordingly” is important”. Let me explain why I don’t agree with this. (Note, I don’t in any way enjoy disagreeing with my colleagues but it’s really important that people get this stuff right.)

1. Once written down to their fair value (i.e the pittance people will pay for them) there is nothing “toxic” about these assets—the banks can keep them, or sell them to private investors, or the government. What’s “toxic” about writing them down to, or selling them at, fair value is that the banks are then insolvent to the tune of €10 billion. That’s why they’re unwilling to sell them at prices investors will offer; that’s why they’re “encumbered” by them.

2. While Plan 1 doesn’t require this, in practice it probably makes sense for the government to buy the underpeforming assets at prevailing market value (i.e. the price that leads the banks to have equity capital of -€10 billion), give the banks some marketable government debt in return, and then use an asset-mangement agency to recoup as much as possible fromt the sale over time of the bad assets. In the scenario in which there is some systemic underpricing of these assets (which, frankly, I doubt) then perhaps the government may end up losing less than the €10 billion. In this example, the arithmetic behind this approach is the exact same as nationalising and leaving the assets on the books at a written-down fair marketable value but it reduces problems associated with Knightian uncertainty and the like.

I think John’s preference share idea is an interesting option but I don’t think it would work. The size of the preference share investment would be such that, most likely, you’re just delaying nationalisation and appropriate restructuring. The banks everywhere know this and are lobbying against taking more preference share investment. And by placing a huge debt-like burden on the banks, you are indeed encouraging gambling for redemption.

On the political economy difficulties of nationalisation, remember that the FDIC regularly goes in and takes over banks, cleans them up and then sells them on to private investors. It’s much less unprecedented than many think.

Karl: Thanks so much for the useful comments. Who says economics is dull! It is great to have at least a bit of controversy — if we all agree it is boring.

Your new proposal does have the warehousing aspect that the bad loan assets are separated from the banks so that the cash flows from new lending is not intermixed with cash flows from old bad loan assets. That intermixing is the reason for the mis-incentives generated by the bad loan assets on the bank balance sheets. It loses the idea of targeted quantitative easing (what someone would call propping up rich bankers) which is a drawback of that plan in terms of impact on credit-liquidity in the long-term debt market.

Nationalizing AIB and BOI seems quite extreme so if it could be avoided perhaps it should be. Also I like the European dimension of a Euro-wide Geithner plan.

In a credit-liquidity crisis, agents’ implicit state prices differ so that a central bank can have a much lower state price for issuing defaultable bonds compared to private sector investors such as hedge funds etc managing these assets. So there is an efficiency gain from letting the ECB provide a liquidity-related defaultable bond. This makes simple arithmetic calculations not reliable if there are differing liquidity premium across agents. That is a bit technical and complicated but I do not think it can be ignored. In a credit-liquidity crisis, the central bank can increase economic welfare by providing credit and/or liquidity.

Thanks Greg.

Indeed, I agree that there are complications surrounding the pricing of the complex assets in crisis-like conditions (less so the development land underlying the Irish bad loans) and that this makes things more complicated than my simple numerical example. Personally, I’m skeptical of the DeLong argument that there’s a pot of money to made from buying these assets at higher than current market price and then just holding to maturity but really who knows. Another complication is the correlated risk caveat I raised in the original post on this make me wonder whether the Geithner scheme really will get the asset prices up as much as they want.

Still, simple numerical examples have their place. Today’s exchange has convinced me I need to clarify my description of the two plans if I’m going to roll it out again. Perhaps I may punish the blog readers with one more go since the less discerning among our readers may have missed today’s Kildare-Dublin-Ontario exchange!

Another benefit of the Euro-Geithner plan is its effect on accounting-related uncertainty in bank asset valuation. Under the plan, the ECB-debt-supported asset management firms purchase the toxic assets of the banks with cash, and leave the banks with conventional low-risk portfolios of commercial banking loans and mortgages. Afdter this balance sheet cleansing, investors have a good sense of true cash flow value of the banks. With this greater clarity in their valuation the banks’ business strategies are affected in terms of hunting for new lending business.

At the moment the valuation of the banks’ assets is very hard to pin down. Irish banks are required under international accounting standards to use the realized loss model for loan valuation. They cannot officially recognize losses until they are realized. They can provide supplemental information but this does not carry the same scrutiny or reliability for investors. So this improved accounting information is another net value increase associated with the plan.

While I have a big reservation about the Geithner scheme, let’s note some important positive features.

In particular, with its reliance on competition between incentivized private investors has some attractive features relative to alternatives such as the insurance scheme recently offered by the British Government to RBS and Lloyds. The British scheme on the involved case-by-case bilateral negotiation behind closed doors, and left a 10 percent coinsurance with the banks. The U.S. scheme also scores by giving the new managers a more high-powered incentive to recover.

And, contrary to what Karl’s example implies, the Geithner scheme does not undertake to paty enough to the banks to make them whole and well-capitalized — that is clearly a misrepresentation.

Unfortunately, though, the built-in incentive it offers through underpriced financing (which Greg has acknowledged) will likely result in the banks getting paid too much for the legacy loans, leading to heavy taxpayer costs down the road.

In other words, despite the transparent competition, the scheme does not provide a rational “price discovery”. (My previous comments referring to option pricing are, I think, the best way of showing this, though I know readers don’t have the patience to work through this aspect).

By underpriced financing I don’t mean that the rate of interest is 100 basis points too low. It might be 2000 basis points too low or more. And on close to a trillion that’s real money.

For the Irish case, then, only some of Geithner’s scheme (the competitive bidding, the transparency, the high-powered incentives to manage the debt well) seem of interest for the Irish policy context. The large subsidized financing I wouldn’t have the Irish Government touch with a barge pole. (Nor would Greg, as he has said)

So, short of European action, for me we are back with the options that have been bandied around among us for months.

@Patrick: I agree with you that the potentially fatal flaw in my Euro-Geithner plan is that it assumes a functioning central bank for the Euro zone, with the willingness and ability to act decisively in the interest of its constituents. If the ECB does nothing then Ireland cannot act on its own. Ireland does not have a sovereign central bank to inject liquidity into the system. So we are stuck with inferior options if that turns out to be the case.

As research economists at least we get a natural experiment. The US implements the Geithner plan and the Euro zone does nothing beyond dragging its feet and allowing some banks and/or national governments to default on obligations and restructure out of bankruptcy. Then we get to see which plan works better. Our great-great-grandchildren, faced with the next similar crisis, can learn from analysis of this dataset.

Suppose the US government decides to lend to the new Geithner vehicles at some ‘cheap’ rate, say 4 or 5%. The subsidy is the difference between this rate and the rate at which today’s private market would fund the same vehicles, which have no expectations of fancy profits, given the competitive auction feature.
Leverage is to be six times, and the vehicle is to bid competitively for impaired mortgage-backed securities of uncertain value. It seems to me, and to Patrick Honohan, that the market rate could be 20%, 30% or even more. It is entirely possible that six times leverage would not be available at all.
The subsidy is therefore enormous, and passes through to the shareholders. Assuming transaction costs and asset recovery rates are similar, it is a far better deal for the taxpayers to nationalise the banks, realise the value of impaired assets, inject equity, and sell them on. Why do people find this complicated?

Colm, maybe I’m wrong, but I’d be surprised if Patrick would concur that the nationalisation option is uncomplicated. First, if the banks are indeed insolvent, the government will have a hole to fill if it is to keep the banks as going concerns no matter which route it chooses. The government imposed very limited losses on creditors with the Anglo nationalisation. If this is the model, then with market equity near zero there isn’t much to be gained for taxpayers in terms of loss sharing by nationalising. Second, my limited knowledge of the empirical evidence tells me it is a challenge to get resolution/asset management institutions right. The FDIC appears to do a good job in the US. But that institutional capital has been built up over years. And, at the risk of sounding churlish, Ireland is not Sweden when it comes to the quality of public sector institutions. This needs to be thought through carefully before putting the politicians in charge of the banks.

Injecting capital into a broken system has yet to be shown to work in the real world.
Who will borrow from the newly scrubbed banks? Why will they borrow as the system continues to deflate? The ECB is not willing to waste money. The Fed created their bubble. The ECB has seen that and is refusing to follow them into the hole they have dug. We know what is going to happen to the US$ over the next months. When it stops falling then will be the time for a stimulus as we will know what industries are capable of surviving in this NWO.
The US failure will have massive ramifications for Ireland as we were a profit centre for their mobile industries. Mobile if they survive. They may seek lower cost centres as having low tax rates only works if there is a profit……
And protectionism is the ultimate refusal of globalls. Once the money runs out, the sleeves get rolled up and things happen.
We are takers, not givers. Think Iceland and stay sane, not easy though.

@John McHale
Our politicians get away with what they have done because of the appointment of civil servants who then owe the pollies. Any and all corruption is possible then. Remember the DIRT mess? I testified that the Revenue had stopped Inspectors from investigating the banks. Not the C&AG, me! Everyone else was giving them a free pass.
FDIC is about to run out of funds.

John, of course nationalising banks, or the Geithner bail-out, or the UK insurance option, are all complicated and involve operational challenges. The choice between them is not simple. The point I was trying to make about the US Geithner scheme is the same as Karl Whelan’s point: it will cost taxpayers more, if it succeeds, than a nationalisation alternative, since the subsidy on the debt component must be large. This is the bit (perhaps the only bit) that is not complicated.

This is what Jeffrey Sachs says about the Geithner plan in todays’ FT:
“The Geithner-Summers plan, officially called the public/private investment programme, is a thinly veiled attempt to transfer up to hundreds of billions of dollars of US taxpayer funds to the commercial banks, by buying toxic assets from the banks at far above their market value. It is dressed up as a market transaction but that is a fig-leaf, since the government will put in 90 per cent or more of the funds and the “price discovery” process is not genuine. It is no surprise that stock market capitalisation of the banks has risen about 50 per cent from the lows of two weeks ago. Taxpayers are the losers, even as they stand on the sidelines cheering the rise of the stock market. It is their money fuelling the rally, yet the banks are the beneficiaries.”

My suggestion is that the EU launch a suitably modified Geithner plan, appropriate to the conditions in troubled Euro-zone banking markets. The favourably priced non-recourse loans are to be provided by the ECB not the Irish exchequer, which only has to provide the matching equity capital (10% of total capital) with payoffs identical to that of the private investors. The benefit for the ECB is that, rather than provide capital later in a sovereign-crisis situation, it can provide capital now in a highly efficient form of targeted quantitative easing, aimed directly at the commercial banking sector.

The purpose of the proposal is to encourage commercial bank lending to SMEs and others to kickstart Euro-area growth, particularly in the case of Ireland.

There are some undesirable redistributive aspects since ECB liquidity goes into recovering bank shares and prevents bankruptcy and/or nationalization of banks. I do not believe that nationalization is politically on the table so the real alternative is do-nothing muddling along.

A range of corporate governance/remuneration controls should be tied into the proposal. I still think that it is a reasonable proposal with lots of positives, despite the harsh criticisms of its redistributive consequences.

Comments are closed.