Safety nets and whatnot

Over on Felix Salmon's blog, I made the following comment:

I think this is exactly right. If you want people to take more risks, you can either:

A. Increase the benefits they get from successful outcomes (which I view as the generally Republican view, such as reducing capital gains taxes or inheritance taxes)

B. Decrease the costs they bear from unsuccessful outcomes

C. Increase the probability of a successful outcome

I think all of these are plausible goals, but in my eyes if you want me to walk a tightrope, making the wire more stable and adding a safety net is going to jack up the probability much more than adding some gold to the pot on the other side.

As Greg Mankiw or Charlie Munger would say, incentives matter.  But there are often a variety of (changes in) incentives that would induce the same (changes in) behavior.  I left at least one alternative out:
D.  You can make the existing situation (e.g., current job, retirement system) more risky

In any event, it's easy to find examples of people arguing that bigger payoffs encourage more risk-taking.  Some interesting quotes from a 2002 article that does it (not sure why the Hawaii Reporter was the first google result, but who cares?):

President George W. Bush renewed his call for permanent repeal of the estate tax on March 19. "It is unfair, patently unfair, for any entrepreneur ... to develop her own business and have that business taxed twice as she tries to leave her assets to whomever she chooses," Bush told a Women's Entrepreneurship Summit in Washington, D.C. "We must make the repeal of the death tax permanent. I call upon Congress to do this immediately."

 "I do not believe the role of government is to create wealth," the president told last Tuesday's group at the Ronald Reagan Building and International Trade Center. "The role of government is to create an environment that encourages risk taking, an environment that facilitates the flow of capital, and an environment in which people can realize their dreams. ... And that's exactly what I intend to do as the President."

So why is it that I see so much push towards increasing payoffs to investment, but relatively little on enhancing the safety nets for failed outcomes?  I mean, it's understandable to think about the successes, but we can't ignore the "risk" in "risk-taking".  If we truly want to encourage people to take more risks, with the belief that risk-taking promotes economic growth, shouldn't we be using a full mix of incentives?  In other words, why so much "Incentive A" from above, but so little "Incentives B & C"?  Especially since, in my personal view, the likelihood that my future huge estate will be taxed upon my death is an infinitely minor reason for me not to strike out on my own.

One thing I find interesting is that "Incentive D" is in accord with Greg Mankiw's argument for negative real interest rates - make the status quo less desirable to push people into more investment.

Also, as I noted on Felix's blog, I recently saw Man On Wire, which I thought was fantastic.  Hat tip to Tyler Cowen for that.

Housing as an investment 2

The discussion on rortybomb continues, so I thought I’d add some more thoughts.

As humans, we are effectively short food and shelter for the rest of our lives, and you could probably add in insurance/medical care.  (I think Mike at Rortybomb made this point a while ago, but I couldn’t easily google it.)  That means that we need to somehow structure our earnings, consumption, and investment to satisfy those expected needs (future payments), bearing in mind that in our later years we’ll be doing very little earning and investing, and lots of consumption.

However, we have a choice on how we satisfy those needs.  One way is to prepay them while we are working, which is what one commenter suggests for our housing needs:

My view of the housing market, and why I think a lot of this anti-housing as an investment is silly, is that it’s a consumption good you will always need to consume. You’re not going to suddenly decide in 2027 that you might do without housing for a while. On that basis alone then it’s quite sensible to store it up for when you have no income, ie in retirement. It’s an almost perfect hedge of a large chunk of your consumption needs.

The commenter is right in one regard – we certainly do need to store up for our future consumption, when we’ll have no income.  However, the choice is not whether to save for the future, but in what vehicle.  Specifically, the above comment seems to argue that we should be prepaying for our housing needs.  Paraphrasing:  “Buy a house now so that you’ll have one when you’re not employed.” 

While I certainly agree with the need to save for future consumption, I don't see any reason that we should prepay our housing needs in the form of a single, non-diversified asset with large transactions costs.  Similarly, I think we can effectively save for our future food needs without filling our basements with cans of vegetables, soup, and spam.  (And beer!  Don’t forget the beer.  Some, especially stouts and barleywines age very well.)

So the choice isn’t save or don’t save for future housing consumption.  But rather, in what form.  Putting aside some very important factors (differences of housing types available for rent or sale, other payments like property taxes, maintenance, possible rent increases), it's an economic tradeoff between two different-looking cash flow streams.  In the same way that annuities can be easily converted into lump sums (and vice versa), it's straightforward to compare renting (effectively a negative annuity) with buying (a one-time lump sum).  It's inappropriate (in my opinion) to imply that buying is somehow different (and better) than renting because of the timing of the payments.

My impression of this debate is that most people arguing for housing as an investment view the purchase of a home as qualitatively better than renting a home (for reasons that are often poorly articulated), while the other side views it as more of a straightforward evaluation of the timing and magnitude of the cash flows.

Housing as an investment

Over on the fantastic rortyblog, Mike gets involved in a conversation on housing.  Specifically, he discusses the notion that housing is a good investment decision, as argued by Adam Ozimek.


I largely agree with Mike's broad point - too often people overvalue the notion that housing is an investment.  My particular bete noir is the idea that "rent is throwing money away" while mortgage payments are building up equity.  I think it's pretty intuitive that the decision to rent versus buy any asset hinges upon the relative cost of doing each.  Historically, it's been a very good idea to buy in most markets, especially considering that labor mobility wasn't as important as it is today and people were able to stay in one place for longer periods of time, avoiding the significant transaction costs associated with moving.


In any event, I fall in Mike's and Felix Salmon's camp of thinking that housing as an investment is generally a bad idea.  But I actually disagree (I think) with one of Mike's comments:

Felix notes: “DanHess and Matt Turner make the point that buying a house is a great way of forcing people to save over the long term.” There are no free lunches of course, and the reason it is a great way of forcing people to save over the long term is that it is incredibly expensive and difficult to get any money out of it.

I think the more straightforward reason is behavioral.  You've gotten people to commit to saving in a way that isn't transparent, so they're not actually aware they're doing it.  It's just that, 30 years later, they get to put a mortgage document on their grill and have a party when they realize how much equity they've built up.

It doesn't seem terribly different (to me, of course) from:

- withholding social security payments involuntarily

- automatic 401(k) enrollment

- the new programs where employees can commit that future raises will go toward retirement contributions (are these just hypothetical?  I feel like I've read about the idea many times, but haven't seen any actual examples)

In all of these cases, it strikes me that saving is made easier because there was something automated about the process, where the individual doesn't feel the "pain" of foregone consumption. 

(You could also point out that the mechanism of withholding income taxes accomplishes the same thing - reducing the public's understanding of how much in tax they actually pay.)

So overall, I actually do think it's a free lunch, in much the same way that a lot of valuable internet content is a free lunch.  Of course there's a cost - individuals truly are foregoing consumption, and internet contributors truly are laboring to create content without compensation.  It's just that in those cases, the people bearing the cost don't seem to mind as much as they probably should.

I think I like this plan

CalculatedRisk describes his view/guess as to the administration's plan.  I hope he's right, because it sounds very reasonable.  Excerpt:
------------------------------------
It sounds like the stress tests could be completed within "weeks" at some banks, and I think 30 days is sufficient for all 18 or so banks with $100 billion in assets.

The banks will probably fall into one of three categories:

1) No additional assistance required. These banks will definitely want this publicized!

2) The banks in between that will need additional capital. This is where the Capital Assistance Program comes in:

Capital Assistance Program: While banks will be encouraged to access private markets to raise any additional capital needed to establish this buffer, a financial institution that has undergone a comprehensive “stress test” will have access to a Treasury provided “capital buffer” to help absorb losses and serve as a bridge to receiving increased private capital. ... Firms will receive a preferred security investment from Treasury in convertible securities that they can convert into common equity if needed to preserve lending in a worse-than-expected economic environment. This convertible preferred security will carry a dividend to be specified later and a conversion price set at a modest discount from the prevailing level of the institution’s stock price as of February 9, 2009.
emphasis added

3) Banks that will need to be nationalized or sold.

------------------------------------------------------

There are still several issues:

1.  How do you announce the results?  Uncertainty is bad and announcing the results piecemeal will cause great uncertainty.  I think it might be better to announce a fixed date upon which the results for all large banks will be announced.  Even then, I can't even imagine the options market behavior leading up to that date.

2.  Getting new money injected can be tricky.  Especially difficult is figuring out what the government should get in return.  I think that aiming for public-private partnerships would work well.  For example, tell firms that need money, "Listen, if you want money, we'd prefer you arrange for private transactions, such as those executed between Berkshire Hathaway and Goldman Sachs, General Electric, and Harley Davidson.  For every dollar you raise privately, the government will be willing to provide up to $10 in exactly the same structure deal."  This works for both liquidity issues and solvency issues.  If a bank isn't able to convince private investors (including the managers) to raise 10% of the needed cash, I think it's hard to make a case that the government should be willing to save it.

3.  How do you nationalize?  I assume this isn't simply a purchase of 100% of the firm's equity at the most recent market price.  Is it just the OTS coming in like they normally do?  Am I overthinking this?

4.  What's happening to the lenders?  We can't have equity holders wiped out while debt holders are made completely whole.  And what about pension obligations?  I don't know what the solution to this is.  I suspect Luigi Zingales would argue for a cram-down, and I largely agree with him.  But I'm not sure I'm considering all the important factors.  For example, what happens if we piss off foreign debt holders?  Is that important?

In any event, these are difficult challenges.  I think, though, that things are moving in much better directions than they were in October of last year.  In terms of the stock market, it seems clear that investors do not like uncertainty, but I'm convinced that the long-term outlook of the U.S. economy is better now than it was in October.   

That means either investors were still too confident in October or investors are too pessimistic now.

Enough with the "Toxic Assets" meme

I don't know how to solve the banking crisis. Let's get that out of the way.

What I do know is that if people are thinking about the problem in the wrong way, they're less likely to come up with a good solution. One of the wrong ways that people are thinking about the problem relates to the notion of "toxic assets".

Here's one example from Fortune:
Don't forget those toxic assets
which includes the following quote:
The trouble is that the toxic assets did not melt away on their own and they contaminate the ability of banks to redress the quality of their balance sheets. Until the banks do so, they stand virtually no chance of returning to more normal lending activity.

BACKGROUND
Banks serve as intermediaries between savers (who have money) and borrowers (who want it). They take money from savers in the form of checking accounts, savings accounts, and CDs. From the bank’s perspective, these are liabilities. In order to generate funds to satisfy the savers (e.g., to pay interest on savings accounts), pay their operating expenses, and earn a return for the owners (shareholders), banks need to invest that money. Sometimes they’ll lend that money directly to borrowers who want to buy a house or a car or go to school. These loans outstanding are assets from the bank’s perspective. As long as the borrowers make their payments as expected (homeowners make their mortgage payments), the bank can pay its employees, satisfy its obligations to the savers, and have money left over to return to shareholders.

However, when the bank’s assets don’t perform as expected, banks run into problems. So when homeowners begin defaulting on their mortgages at higher-than-expected rates, banks’ assets are worth less than expected, and they have difficulty satisfying their obligations. Right now, banks are having a problem because their assets (money they have lent to borrowers) aren’t worth enough to satisfy their obligations. So that’s the big issue. BANKS’ ASSETS AREN’T SUFFICIENT TO SATISFY THEIR OBLIGATIONS (which effectively means meeting regulatory requirements that asset values have to exceed liabilities by a certain amount).

WHAT IS CAUSING THE DECLINE IN ASSET VALUE?
Now there’s some argument about why this is the case. In short, assets could:
1. Be worth a very small amount because, over the life of those assets (i.e., the term of the loans), the borrowers simply don’t make nearly the dollar amount of payments that banks expected. So a bank that simply holds on to its mortgages for the entire 30-year life gets far less than what was originally borrowed. Consider this a “real” or “permanent” decline in value because it reflects the fact that, in hindsight, banks made stupid loans and they’re going to lose money on those loans.

2. Have a low “market price”. That is, certain assets the bank holds (like investments in pools of mortgages) may actually be traded on an open market. The value of those assets on any given day is based on the observed market price of the asset (just as the value of my assets is based on the quoted prices I see in my brokerage account). Those market prices, in turn, are based on the market’s expectations of what the assets will generate in terms of future cash flows. So the market is trying to estimate, for a given pool of mortgages, say, how many of those mortgages will be paid early, how many of those mortgages will be satisfied over the contractual life, and how many of the mortgages will go in default because the homeowner doesn’t pay. Furthermore, the market has to estimate, for the expected defaults, how much of the mortgage they’ll recover through foreclosure or short sale.

Generally, markets are quite good at estimating the fair value of assets, even very complicated assets. But sometimes there is so much uncertainty that investors simply give up. That is, a bank may have an asset ABC that in 2006 was worth $100,000 based on observed market pricing. But now, investors simply have no idea how much the asset is worth, because they have no idea how high default rates are going to be, or how high recovery rates on foreclosed properties will be. So they say something like, “I wouldn’t be willing to pay more than $5,000 for that asset. But I wouldn’t sell it for less than $95,000. I simply have no idea how much it’s worth and I’m not going to risk money on something with so much uncertainty.”

So now you’ve got banks holding on to assets with highly uncertain values. They don’t want to sell the assets because no one is willing to buy them at what the bank deems a fair price. In the prior example, you might say the expected value of the asset is $50,000 based on the midpoint of the two estimates, but if the bank sold it they’d only get $5,000. These assets with highly uncertain values are being identified as “toxic assets”. And the argument that I see in stories like the Fortune article is that if you can simply remove these toxic assets from the bank’s Balance Sheet, the bank will be perfectly fine.

SO WHAT’S THE PROBLEM?
The argument doesn’t make any sense. The banks aren’t harmed by having those toxic assets on their Balance Sheets. The assets aren’t toxic in the sense that they will actually contaminate any other assets in close proximity. The banks are harmed because they have toxic assets and insufficient other, non-toxic assets to satisfy their obligations.
As an analogy, suppose I borrow some money and use that money to invest in baseball cards. In particular, I use all of the borrowed funds to buy a large supply of 1968 Johnny Bench rookie cards.

Subsequently, it turns out that nobody wants Johnny Bench rookie cards anymore – I can’t sell them, except at super-distressed prices. So I’m stuck. I’ve got to make payments on the loan and all I have are these toxic baseball cards. BUT SIMPLY REMOVING THESE CARDS FROM MY BALANCE SHEET DOES ME NO GOOD. If I burned these baseball cards, I still have no other assets and I still owe money. I can’t be better off by removing the assets from my Balance Sheet. The real issue is that I need some other assets with which I can satisfy my obligations.

The same is true for banks. They don’t benefit by simply removing the toxic assets from the Balance Sheet. They need other, more liquid assets. Framing the problem in any other way is simply obfuscating the central point.

THE SOLUTION
The end result is that troubled banks will either receive capital, be allowed to survive with reduced regulatory requirements, or fail. Regardless of how you feel about letting banks fail in a structured fashion or letting them continue with reduced capital requirements (I think there are merits to both), if banks are going to receive capital from the government, the primary question is “what does the government get in return”. Does the government get shares of equity (common or preferred), does it get debt (and if so, what is it senior to?), does it get the “toxic” assets (i.e., a straight asset sale), or is it simply an injection with no strings attached?

Determining the right structure is difficult for many reasons, even putting aside the dollar amounts (e.g., should the government have voting rights, should the government be allowed to jump ahead of senior claimants in terms of debt priority?). But the dollar amounts are difficult for precisely the reason that these assets are “toxic” to begin with – nobody really has any idea how much they’re worth. So whether it’s a loan or an equity injection or an asset purchase, the question is going to be “how much does the government get in return?”

I don’t know the right way to do this. But I do know that solutions posing as “removing the toxic assets from the Balance Sheet” aren’t doing anyone any favors. If those arguing for removal really mean converting the illiquid assets into cash, they should say so. And they should provide the mechanism by which the dollar amount should be determined. And they should indicate why an asset sale is better than a debt or equity injection. (The market in banks’ common equity is still liquid, so at least the government is more likely to get a fairly-determined price if they were to purchase equity compared to what they’d get if they purchased illiquid asset-backed securities.)

And if those arguing for removal actually believe that simply removing the assets from the Balance Sheet will solve the problem, they should be ignored. And they should send all their unwanted assets to me.