Two Harvard professors — John C. Coates from the Law School and David S. Scharfstein from the Business School — had an op-ed in the NYT yesterday in which they identified a basic design flaw in TARP. Their criticism is simple and cogent, and doesn’t require any background in economics to understand, yet it has received surprisingly little attention. Which is troubling because their criticism has implications for how the new Financial Stability Plan should be structured.

For reasons that remain unclear, the Troubled Asset Relief Program has channeled aid to bank holding companies rather than banks. The Obama administration’s new Financial Stability Plan will have more influence on bank lending if it actually directs its support to banks.

To see why, it’s important to understand the distinction between banks and bank holding companies. Banks take deposits and make loans to consumers and corporations. Bank holding companies own or control these banks. The big holding companies also own other businesses, including ones that execute trades both on their clients’ behalf and for themselves.

It would seem obvious that helping banks, not holding companies, would be the most direct way to stimulate bank lending. But when TARP purchased preferred stock and warrants, it bought them from holding companies, not their bank subsidiaries.

While TARP has been generous with bank holding companies, these companies have not been so generous with their banks. Four large holding companies — JP Morgan, Citigroup, Bank of America and Wells Fargo — initially received a total of $90 billion in TARP money in the fall, but by the end of 2008 they had contributed less than $15 billion in equity capital to their subsidiary banks.

The holding companies seem to have invested most of their TARP money in their other businesses or else retained the option to do so by keeping it in deposit accounts, even as the capital of their banks decreased. At the same time the banks, which provide the majority of loans to large corporate borrowers, drastically reduced lending to new borrowers.

It’s easy to see why holding companies would withhold capital from their troubled banks. If a bank is insolvent — as many are now believed to be — and the government has to take it over, the holding company loses any capital it gave to the bank. Rather than take that risk, the holding company can opt to spend its money elsewhere…
[...]
Bailouts, at the very least, should reach their target. When Washington wanted to help Chrysler, it gave money to Chrysler. It did not write a blank check to Cerberus, the private equity firm that owns Chrysler, in the hope that the money would somehow find its way to the carmaker and not to the other companies Cerberus owns.
[...]
It would be more effective to simply ensure that the Financial Stability Plan is directed at banks. When the government buys stock, it should buy bank stock. And if it chooses to buy stock in holding companies, it should at least require that the new capital reaches the bank and non-bank subsidiaries that the government wishes to support. If the government chooses to help private investors buy toxic bank assets, as the planned Public-Private Investment Fund is supposed to do, it should not allow the banks to send those investments to their holding companies. And if the government decides to guarantee debt, it should guarantee the debt of banks, not of holding companies.

Once they lay it out like that, it’s self-evident that it made absolutely no sense to operate TARP this way. Unfortunately, it’s too late to do anything about TARP. Or to hold anyone accountable for such obvious stupidity.

But I sure as hell hope that someone who is in a position to influence how the Financial Stability Plan will work is listening. We really can’t afford déjà vu all over again.