David R. Henderson  

Alan Blinder: Banks Don't Pay Taxes

The Magic Solution of Devaluat... What Did You Learn in Business...

Oh, and, by the way, the federal government is a taxpayer.

In an op/ed in Monday's Wall Street Journal, "How Bernanke Can Get Banks Lending Again," Princeton University economist Alan Blinder writes:

But suppose it doesn't work. Suppose the Fed cuts the IOER [interest on excess reserves] from 25 basis points to minus 25 basis points, and banks don't lend one penny more. In that case, the Fed stops paying banks almost $4 billion a year in interest and, instead, starts collecting roughly equal fees from banks. That would be almost an $8 billion swing from banks to taxpayers. There are worse things.

Now, before you jump on me for disagreeing with Alan Blinder's proposal, please understand that I think it could be a good one. I've long said that one of the most exquisitely bad cases of timing was Bernanke's starting to pay interest on reserves when what is needed is an increase in the money supply. I'm uncomfortable about taxing reserves, as Blinder wants to do, but certainly cutting the IOER to zero makes some sense.

My objection, rather, is to Blinder's strange use of language in his second last sentence. In saying that the tax is a swing from banks to taxpayers, Blinder is saying:
1. Banks that pay taxes aren't taxpayers, and
2. When banks pay taxes, the government, which collects taxes, is a taxpayer.

But, pretty clearly, banks do pay taxes. Blinder must know this because in the same article, he advocates taxing them. If you pay taxes, you're a taxpayer.

Next, Blinder thinks that when the taxes are paid by banks, they're given to other taxpayers. They're not. The entity that gets these taxes is the government, not taxpayers. Now he could claim that when the government gets the $4 billion in added taxes, it cuts other taxes by $4 billion. I doubt that he believes that. He could claim that it cuts the deficit by $4 billion and, therefore, the present value of future taxes is $4 billion less. There he could be on stronger ground. But there's a lot of public choice analysis involved in figuring out the right answer about future taxes. He would be better off having written what I'm pretty sure he knows and it would have gone something like this:

In that case, the Fed stops paying banks almost $4 billion a year in interest and, instead, starts collecting roughly equal taxes from banks. That would be a payment of $4 billion less to banks and a $4 billion increase in taxes on banks. There are worse things.

It doesn't quite have the same ring, does it? It might get people wondering whether banks should be taxed more and whether government should get more taxes.

HT to Charley Hooper.

Comments and Sharing


COMMENTS (6 to date)
Silas Barta writes:

Most of these articles seemed to be premised on the idea that more bank lending is necessarily good. But it's only good if it's invested in economically viable projects. Bank loans would *not* be good if they were only made because they were effectively forced to.

Loans are not good. Good loans are good. (Sorry if this is slightly OT)

Bill Woolsey writes:


I am glad that you consider it not to be a bad idea, and are simply making a point about banks being taxpayers too.

However, I think it is better to think of this as a storage fee. Banks pay a fee for services provided by the Fed, making the taxpayers money.

Since banks aren't required to hold excess reserves, if they don't want to pay this fee, they just have to do something else with the funds. They might make commercial loans, though they could "lend" by purchasing already existing bonds--even government bonds. If this reduces the interest rate on government bonds, the taxpayers have a savings.

Realistically, the banks would reduce the interest rate they pay on deposits--almost entirely government insured deposits. And so, depositors would pay a "storage fees" for holding money in banks. Bank depositors can, of course, avoid this and earn a yield by buying corporate bonds or stocks. Earning a yield would require bearing risk. Having the government guarantee your wealth would cost. Or, of course, they could spending the money on consumer goods. Businesses holding money in bank deposits could spend on capital goods.

Normally, of course, people are paid to put money in banks. And banks don't just "store" fund at the Fed. They instead have lots of loan opportunities.

And from the Fed's point of view, issuing reserves that pay a little interest, and holding short and safe government bonds that pay a little more interest, makes money. But now we have a situation where the Fed is bitching and moaning about buying assets and expanding reserves. It is obviously something they consider "costly." And so, they should charge banks a cost for "storing" their money at the Fed. And banks should pass that on and charge people for storing their money in government insured deposit accounts.

You know, when you get a service from the government (like storing your money) it is usually called a user fee.

SheetWise writes:

These contradictions should be pointed out more often.

In political rhetoric the State assumes the role of the people (the taxpayers), the peacekeeper, the fountainhead of wealth, the source of all good, and other heroic roles whenever it is convenient -- but rarely does it recognize or assume its simple role as agent and caretaker.

MG writes:

Clearly, you have spotted Blinder's ideological blind spots. But is the idea transformative, or even worthy of being given a good test? What is the elasticity of supply of risky lending to a change in 25 bp in interest on riskless reserves? Can it possibly explain, let alone account for, the so-called lending gap (actual against some standard)? Risky lending premiums (returns on capital) may not currently be high enough to entice more lending (uncertainty about value of future cash flows, value of collateral, etc)? And of course, demand factors may be also depressing lending. In this case, reverting to the original scheme of paying no interest on reserves may do little. The temptation will then be to exact higher opportunity costs through more tax tweaking. I think the risks of abusing the power to tax (which we have just been reminded is pretty darn unlimited) and the unintended consequences of bank's complying or avoiding this tax would outweight the intended benefits. A lot of the policy talk that has come out of DC since 2009 revolves around greasing the prime bank lending lending machine while gumming up many other lending channels/enablers (sub prime, securitization, money market funds, calls for higher bank capitalization). This seems curious.

Michael Rulle writes:

As Arnold pointed out a while ago, Blinder and Zandi used their famous "1.7234" multiplier (or whatever) to forecast the results of the stimulus legislation. When the prediction failed, they were quite sanguine with the results. Since the model had to be correct, other factors were altered to match history, so they could demonstrate that "X" number of jobs were "created or saved". So Blinder's language, perhaps, should not be surprising.

Re: Excess reserves. I am beginning to ask the question: "who says they are excess reserves"? They are excess in the sense that regulatory bodies say so. But given that banks are not leveraging them, perhaps they are not really "excess" at all. The banks must feel it is too risky and the reserves are not in the least excess.

I admit to reading a lot of Scott Sumner lately. But, again, why do we really think these reserves should be called "excess"?

Justin R. writes:

"Since banks aren't required to hold excess reserves, if they don't want to pay this fee, they just have to do something else with the funds."

If there was a -.25% IOR, it might become viable for banks to request that their reserves be paid out in currency, i.e. the storage cost of currency might be less than the storage cost of reserves.

Comments for this entry have been closed
Return to top