Arnold Kling  

Limited Liability and Banking

Happy Birthday, Eugen and Juli... The Optimal Scapegoat...

Tyler Cowen says that they do not go well together.

If a shareholder invests a dollar in a big bank, why not make that shareholder liable for the first $1.50 -- or more -- of losses as insolvency approaches? In essence, we would be making the shareholders liable for the costs that bank failures impose on society, and making the banks sort out the right mixes of activities and risks.

This sounds good in principle, but I wonder how it would work in practice. Suppose that you sell your shares in Shakee Bank to me today, and tomorrow Shakee has to be taken over by the FDIC. Am I liable for losses, which probably were caused by decisions made before I bought my shares? Suppose that Shakee has accumulated $8 billion in losses, and all its shareholders of record obtained their shares for a collective $0.10. What happens then?

Without more explanation, this idea seems even less workable than my proposal to increase the liability of managers of failed banks by putting them in jail.

Comments and Sharing

COMMENTS (24 to date)
Yancey Ward writes:

Or what is the justification for not making the shareholders of record responsible for the GM and Chrysler bailouts at the time? What is special about banking in this instance?

Let's Be Free writes:

I would stop investing in banks immediately and expect others would as well, making this at its core a proposal for eliminating a privately owned, privately capitalized, competetive banking system, setting the stage for the government to step in with a monolithic national bank, making all investment and lending decisions and defining the terms on which all moeny is saved. Good luck with that.

Daniel Artz writes:

Putting shareholders at risk for more than their investment is contrary to the very essence of the corporate form (unless, of course, you can demonstrate the facts to support one of the common law causes of action for piercing the corporate veil). However, I would strongly support a Federal statutory form of "fraudulent transfer" recovery from managers and directors. Set a 6 year look back period, and provide that any officer, director, or manager who received any bonus based upon profits that turned out to be illusory, or anyone who received total compensation in excess of a statutory "safe harbor" (I propose $300,000 per annum, but I'm open to suggestion) is subject to liability to recover all such excess bonuses or excess compensation in the event of any bank failure resulting in any loss to depositors or creditors. As in Bankruptcy, make the Receiver (the FDIC, in the case of Federal or Federally-insured banks) responsible for bringing actions seeking recovery under this statute. That way, the people who are responsible for making the risk-based decisions, and who stand to benefit from bonuses based on the transactions at issue, are put at risk for at least a portion of the losses. This would better align financial incentives of the bank managers with the public interest.

Æternitatis writes:

Prof. Kling's argument against the idea assumes extreme myopia on the part of investors. If a bank is close to insolvency (and even more so if it is $8 billion beyond insolvent) tomorrow, investors today at least have an inkling of the risk. In other words, investors today would not accept the shares of Shakee Bank unless they are paid a negative price to do so. As long as shareholders of record can be identified and you cannot just drop your shares in the dumpster, there is no reason to assume that a stock market couldn't function with negative prices.

An alternative would be to bundle every share of the bank with a special permanent bond that (a) pays, let's say, the fed funds rate and (b) is first in line to take losses after the equity but before any other bond holders. This would accomplish roughly the same as Prof. Cowen's proposal, but would not require negative prices. It might also be more acceptable to investors as it caps their risk at the value of the bond (which would allow fine-tuning).

Æternitatis writes:

I should add that there is one problem with negative prices that I'm unsure how to solve: If I held a bank share with a negative value of one billion, I would be very tempted to sell my entire holdings for a very small negative price (-$100, let's say) to somebody who is already insolvent or nearly so (the bum around the corner, for example). Without tracing liability back to previous owners if the current owner is insolvent (which would create the unfortunate situation that you could never really clear yourself of liability if you ever held a bank share), I'm unsure how to fill this gap in the negative price idea.

steve writes:

As Cowen notes in his piece, banks are good at hiding stuff off their balance sheets. Shareholders will have to make uninformed guesses about value, knowing they might be on the hook. My guess, is that they assume the bank will still get bailed out.

There is also a timing problem here. It will take months to sort out true shareholder value. Much better to quickly resolve the bank and get credit markets functioning again. Best long term solution is to align incentives. Get rid of LLCs and make banks function as partnerships, with partners having full liability. Up to and including loss of personal assets. I think at that point. they will start listening to risk management.


Phil writes:

You don't really need to tie it to the price paid for the shares, do you?

To every share, attach an obligation that requires you to pay a fixed $X per share you own if the bank becomes insolvent. (That will push the price of the share down somewhat, but a lot less than $X if the bank is doing fine for now.)

The brokerage will require you to have enough margin to cover the $X, just as they require you to have a certain margin when you're short a put option.

If insolvency becomes likely, the shares might actually have a *negative* value. But that's not a big problem ... just means you have to pay someone to take them off your hands.

But, you know, there's an easier way to do this. Just require the bank to permanently hand $X per share over to the government. Every year, the government pays shareholders a dividend on interest earned on the $X. If the bank goes insolvent, the $X reverts to the government.

Almost the same incentives ... the only problem is that the shareholder has to tie up $X of his capital, instead of just guaranteeing $X. I suppose the government could fix that, if it really wanted to, by subsidizing brokerages to lend the $X back to the (creditworthy) investor at the same rate the government is paying on the $X.

Greg writes:

its a good idea but i think there needs to be some more thinking of why make it $1.50 and not for example $1.60.

i think this would make investors into banks much more cautious (as in they will not invest unless they are absolutely sure of what the banks are doing). although in the short term the value of the shares will sharply fall as it will make holding shares in banks very risky. it would take time for shareholders to get to the right equilibrium of how much they know about the bank they are investing into and the costs of finding this information.

you also have to consider the effect on financial innovation: without limited liability banks would not be so keen to create new investments strategies. but i think this is the general idea, we want to make financial institutions more cautious i think.

Joe Cushing writes:

I seem to remember that there was a rule limiting the power of institutional investors in banks. A small investor votes by buying or selling shares but an institutional investor can vote by voting for board members. I think the limit is on the percentage of ownership that any one entity can have. If there were more institutional investors, there would be more accountability.

andy writes:

Double-liability was enforced in the USA until 1930s. So, maybe you should look into the history, how this was enforced?

andy writes:

And I would add that Scottland (except 2 public banks) had ulimited liability until the late 19th century.There was one bank crash which was characterized by shareholders claiming they weren't shareholders :) Anyway, the scottish system seemed to be quite unregulated and quite stable.

Mark V Anderson writes:

If we want to make the bank owners liable for debts of the bank, simply get rid of limited liability. Require all banks to be partnerships instead of corporations. Partners are liable for the debts of the partnership. Perhaps one would allow some limited partners, but the general partners would still be liable. Or maybe disallow limited partners also.

I don't know if partnership banks could raise the level of capital needed to run a bank that would be able to handle the financial needs of multinationals. I suspect not, but just as likely as the hybrid corporation suggested by Tyler. And much simpler than trying to impose debts on shareholders.

Adrian E. Tschoegl writes:

As andi points out, this has been tried and found wanting. In the 19th Century we saw both unlimited liability and shares sold below par (say half par), with the remainder being callable. In 1851 the Banker's Almanac wrote:

The Times says the regulations stipulated with regard to its issues are sufficiently in accordance with the sound views that now prevail, and although the provision that the proprietors shall be liable for twice the amount of their subscription shows that the Government has not yet learned the fallacious nature of all merely hypothetical responsibilities, which often fail when they are really called upon, while at the same time they mislead the public and lessen the caution that would otherwise be exercised; the error is one which under the ordinary routine ideas will most probably be regarded with favor rather than disapproval.

PrometheeFeu writes:

Seems to me there is an easy way to go about that without actually legislating in this manner. Eliminate the FDIC. I'm sure people would take a second look at banks with limited liability structures if they knew the government wouldn't bail out depositors.

Unindicted Co-conspirator writes:

Wouldn't Coase's answer to this be that the risk of being on the hook for the liability was reflected in the price you paid for each share?

Or did I misunderstand something along the way?

"Suppose that you sell your shares in Shakee Bank to me today, and tomorrow Shakee has to be taken over by the FDIC."

I do not agree that government HAS TO take over a bank, any more than it HAS TO police domestic disputes in foreign nations or it HAS TO assure everyone is covered by medical insurance. People working within the state respond to the incentives in their careers and CHOOSE TO do these things.

Remember that limited liability itself is an act of state. Under limited liability the state promises, in specific circumstances, to insert its police between people in a dispute. The state promises to protect people who have made promises from the people to whom they have made promises.

The certainty that this protection can be found changes the atmosphere in many businesses, I believe. Life would go on in better ways if this whole act of state were repealed.

Blake Johnson writes:

Arnold I think the problem is not so great as you imagine it. When you purchase the shares from me, you will be incentivized in light of the increased liability to be more cautious in your choice of whether or not to purchase. Thus you will attempt to take into account to the best of your knowledge the potential that the company will go under after you purchase it.

The adoption of limited liability was largely done to reduce these information costs and lower the cost of capital for joint stock companies. However, there have been instances in the past of companies with double, triple, or even unlimited liability.

For instance, in his work on Free Banking in Scotland Lawrence H. White discusses the fact that many shareholders had unlimited liability. This kind of greater liability system causes investors to keep a closer eye on their companies because of the greater risk it poses to them. Many of these unlimited liability companies lasted for quite a long period, as their conduct convinced investors that they would not engage in the kind of activities that would put them at risk. Tyler also linked to Scott Sumner's blog post on this subject, which itself linked to a paper discussing the use of relatively less limited financing schemes in the U.S. during the National banking era which discusses the kind of systems markets set up to adapt to the problem you discuss.

Thomas Esmond Knox writes:

If one distinguishes between "Paid Up Capital" and the price one pays for the share in the share market, it can readily be seen that in the event of a bank going bust one will be losing much more money than the "Paid Up Capital" value of the shares.

So the market has already solved the problem.

Next question?

Matt Shulman writes:

I think this is equivalent to saying the shares cost $1.50.

Saying the shares cost $1 and the investor is liable for another 50 cents, is really just a convoluted way of saying that the shares cost $1.50. The only difference is that when each share that is purchased for $1.50 it also includes the option to borrow up to 50 cents interest-free from the bank if you like.

I don't see why it would be useful or efficient to bundle an optional interest-free loan with the sale of each share of bank common stock.

Jeremy, Alabama writes:

It would be better to put a portion of distributed profits into escrow for so many years.

But this whole thread is dripping with liberal belief in forcing an outcome, without concern for unanticipated consequences. But the unintended consequences are everything. We are discussing changing liability structures to force 'approved' behavior onto banks. Here are some consequences:

- banks must now buy even greater political protection

- if liability rules are very onerous, ownership will be laundered through offshore corporations, making them ever more unregulatable

These people have billions at stake, and they are smarter than you. When you say you have the power to do something, then you are asserting that the power exists, and you have added that power to the weapons available to the people who already have billions.

Big money is going to do what it is going to do. The absolute best you can hope for is transparency. Onerous rules ensure that will not happen. Call me old-fashioned, but the only way to enforce proper market discipline onto banks is to let them fail, the bigger the better.

Daniel Artz writes:

Just why are the posters on this topic so intent on finding a way to punish, or create anti-risk incentives for, shareholders? One of the defining characteristics of the corporate form is a division between ownership and control. If risky behavior by banks is the problem, then put the incentives on those with the power to change that behavior; the officers, directors, and managers, not the shareholders. Of course, risky behavior by banks becomes much less of a public problem if we just do away with the entire notion of "too big to fail" in the first place, and either eliminate Federal Deposit Insurance, or at least impose risk-rated premiums for that insurance, so that as a bank becomes more unstable, it's cost of maintaining FDIC insurance increases substantially.

Andrew Wallen writes:

The attached link leads to a paper on the trading of unlimited liability bank shares from the 1820s through 1879 in the UK.

This seems like a possible, market based solution. My guess is that people would realize that banking carries a large down-side risk (similar to selling short on a stock). The result could mean banks become smaller and less capitalized when compared with banks today because fewer people are willing to accept the risk.

[comment submitted Feb. 12; restored after accidentally being deleted from our spam folder. We apologize for the inconvenience. -- Econlib Ed.]

Floccina writes:

I think that the current system has feedback problems that make it tough to remain stable but making corporations loose their limit liability status slowly and progressively as leverage goes up might be a plus but it would have to be a complicated to cover all possibilities and avoid periodic running for the exits.

Better IMO to work on the feedback problems so that banks are strengthened by the failure of their competitors (and are encouraged to lend more and spend more). If banks are measured against other banks only the weakest banks could ever fail. With he current system all banks can all fail at once as they can be out competed by Government bills and bonds. I think some free banking schemes would work this way.

diomavro writes:

Although I agree that limited liability doesn't fit banking I don't think that a fixed loss is really a good idea. Just make it unlimited liability, if it was a fixed loss per share, then companies would just buy back their shares to ensure their investors have shares worth more and are liable for less loss.

Comments for this entry have been closed
Return to top