As the Europeans busily discuss how to provide bailouts from the "European Stability Mechanism" directly to banks, Jyrki Katainen, the prime minister of Finland (a country whose government has triple-A bond ratings), has asserted that the "mind-set" must be changed "from bail-out to bail-in."

When it comes to sharing in the losses of failing banks, Katainen asserts, in particular, that shareholders and bondholders should take losses – absolutely right — and that "only in rare, exceptional occasions, public money should be used" —again right, although the historical record demonstrates that such occasions are not a rare as one might hope. 

But Katainen did not mention depositors. What about them? Depositors are lenders to banks by another name. A remarkable feature of virtually all discussions of banking crises these days is that they simply assume depositors should always be protected, and that money should be taken from taxpayers to give to depositors to enforce this proposition, if need be. In the U.S., large depositors are theoretically at risk, but usually the Federal Deposit Insurance Corp. arranges mergers of failed banks so that all depositors are, in fact, protected. Moreover, in the financial crisis of the 2000s, all size limits for deposit insurance were increased and the FDIC model was widely praised.

This is strikingly different from the aftermath of the crisis of the 1980s, when many analysts of the disaster pointed to the perverse effects of government deposit insurance. So, while we are about addressing who should be sharing losses, let's consider whether at least some depositors should be bailed-in, too.

If people buy stock, its price can go down, even to zero. If they buy a bond, the company can default and they may get 50 cents or less on the dollar. If they buy a house, its price can go down a lot — they can lose all the equity they have in it. If they buy gold, its price can drop. Ordinary people can and do take such risks. But if they buy a bank deposit, it is assumed that the government must make their claim risk-free, that they must never get less than 100 cents on the dollar, even when this imposes costs on other, innocent citizens. The global mind-set is certainly set on bail-out, not bail-in, when it comes to depositors.

Let's review the logic of this stance. Demand and other transactions deposits represent the bulk of what is used as money to settle payments in a modern economy. There is a reasonable argument that "money" should by definition always be worth par. Let us grant this argument. Transactions deposits today represent 13% of total deposits at U. S. banks, according to the Federal Deposit Insurance Corp. OK, these should always be worth par. But what about the other 87% of bank deposits? Why not bail them in?

Another possible exception: How about preserving the classic idea of small savings and thus savings accounts, conjuring up as they do visions of Jimmy Stewart and the virtues of thrift? OK, let's exempt these from sharing in losses, too. Savings deposits represent another 17% of U.S. deposits. Between transactions accounts and savings accounts, we have 30% of deposits.

Now: what about the other 70%: the nontransactions, nonsavings deposits? What are they? They are simply fixed-income investments made for a yield, in competition with lots of other notes and bonds. These deposits include time deposits, certificates of deposit, money market deposit accounts and notably, brokered CDs.  As fixed-income investments, they are an excellent candidate to join other fixed-income investments, i.e., bonds, issued by banks, as creditors to be bailed in, consistent with Katainen's ideas.

Because they are simply fixed-income investments with a different name, in a reformed world, 70% of deposits might rank pari passu with senior bonds when it comes to sharing losses. Obviously, this makes them junior to transactions accounts and savings accounts, and also junior (instead of senior, as they now are) to taxpayers.

Such a bail-in would substantially reduce the perverse incentives and moral hazard of government deposit insurance, while arguably being more true to its original purpose. Buyers of these investments would care a lot more about the credit standing and capital of the bank and need appropriate disclosures, just like buyers of other investments sold to the public. Deposit insurance would have to be restructured.  Banking leverage would fall and capital ratios would rise. People would work on how to game the system, of course, but moral hazard in banking would be much reduced.

Whatever one may think about the details, we should reject the current article of faith and instead consider which depositors should be bailed in along with bondholders.

Alex J. Pollock is a resident fellow at the American Enterprise Institute in Washington. He was a president and CEO of the Federal Home Loan Bank of Chicago from 1991 to 2004.