Posted on

Gold vs. US Treasuries

John Hempton writes:

We live in a strange world – the 10 year US Treasury is trading with a 2.63 percent yield.  The market is presuming that there will not be much inflation in those ten years.  However if there is deflation (as per Japan) then the 10 year will wind up being a very good investment (see my blog post on Japanese bond yields from the perspective of a Japanese household).

At the same time gold is appreciating very sharply – from $950 per oz to $1250 in the past year – and from $800 two years ago or $450 five years ago.  On the face of it the gold price is predicting inflation.

Try as I may – I can’t see any reason why both those prices are correct.  I have long held the view that prices are mostly sort-of-rational … [s]o either there is a theoretical way in which both these prices can be correct or even my weak version of the efficient market hypothesis is spectacularly wrong.

and then asks

My first question thus is can anyone tell me why these prices could possibly be consistent?  Is there a rational reason why the bond market is pricing low inflation and the gold market seemingly pricing high inflation?  Does anybody have the ingenious world view in which both these prices are correct?

Since Blogger rejected my comment over at John’s site as being too long, I may as well reproduce it here. I don’t know about “correct” and I’m no finance guy, so my first point is that  I have no freakin’ clue.  Nevertheless, here are five, somewhat contradictory ideas, three of which might fit in a weak EMH world …

Idea #1) Yes, yes, your whole post was predicated on some weak version of the EMH. However … Treasuries, despite what the arch-conservatives are saying, are unlikely to be in a bubble (see idea #4 below).  It might (and only might!) even be impossible for them to be in a bubble.  On the other hand, gold can experience a bubble (to the extent that you concede that bubbles can exist at all).  Just because it can doesn’t mean that it currently is in one, but if it is and treasuries are not, that would partially resolve your dilemma.

Idea #2) Gold, as a commodity, is a affected by global phenomena, whereas US treasuries, while obviously still influenced by global pressures, are more sensitive to the US economy than is gold.  This statement will become more true over time as the US economy shrinks as a share of global GDP.  Therefore, perhaps you should deduce that markets are predicting low inflation or deflation for America, but quite high inflation for the world as a whole.

Idea #3) Gold, as a commodity, partially co-moves with other commodities, many of which are seeing price increases because of real, observable events in their markets (Chinese construction, Russian drought, etc).  Perhaps it is being dragged up by those (this augments idea #2).

Idea #4) In the broad market for USD-denominated investment-grade bonds, there has, I believe, been a net contraction in supply despite the surge in US government borrowing.  This is the private-sector balance-sheet correction.  One might argue, from something of a monetarist point of view, that (disin|de)flation is occurring in the US precisely because the US government is not expanding its borrowing fast enough to replace the private-sector contraction.  I mentioned this briefly the other day.

Idea #5) Another non-EMH idea, I’m afraid:  Both the USD and gold enjoy safe-haven status.  An increase in generalised fear (Knightian uncertainty, unknown unknowns, etc) will shift out the demand for both at all price levels.  To the extent that such a dynamic exists, I suspect that it ebbs away only slowly and, while elevated, is susceptible to rapid increases in response to events that would, in normal times, not affect people so much.

Update 11 Oct 2010:

James Hamilton on essentially the same topic.

Posted on

Don’t put a nappy on me just because you’re mollycoddling idiot #8,749

I hereby present the latest iteration of my telling the world how it ought to be run, damn it.  The topic today is:

With (very low) probability, p, event X might occur at location Y, causing offence, harm or even death to a person of type Z.

There are many examples of this sort of scenario.  Here are a few:

  • X:  Fall off a swing
  • Y: A public park
  • Z: A small child
  • X: Trip and fall
  • Y: Footpaths (sidewalks) with cracks in the concrete
  • Z: Old, clumsy or spatially unaware people
  • X: Bringing your dog
  • Y: The outside seating area of a cafe
  • Z: People that are allergic to, or just have an aversion to, dogs

Here is another, eloquently opposed by M.S. at one of The Economist‘s blogs:

  • X: Drown
  • Y: Lakes in Massachusetts state parks
  • Z: A weak swimmer

I’m sure that you, my eager and most imaginative audience, can describe any number of other examples.

What should we do when confronted with these scenarios?  Most people think that the world positions itself along a line separating, at one end, complete government regulation and at the other, zero government regulation and, instead, the use of tort law and civil suits to restrict the “bad” behaviour.  This is poor logic, however, because it is overly simplistic; it presupposes that we need to do anything at all!

I favour a midway point between regulation and tort law, but more importantly, to my mind, we usually don’t need to do anything when confronted with these possibilities. Life inherently has risks and, while we should act to avoid exacerbating those risks, we should not necessarily seek to remove them altogether.  There are three reasons for this:  First, because removing all risk is simply impossible and it is usually the case that reducing risk in one area causes it to rise in another; second, because exposure to some risk is crucial in the development of well-adjusted people and a properly-functioning society; and third, because to restrict people’s choices in order to lower the risk they face is to deprive them of their basic liberty to choose whether to accept that risk for themselves.

Let me summarise my view in this not-even-remotely-to-scale little plot. Think of the horizontal axis as a measure of how easy it is to demonstrate harm to a point of warranting action by “the authorities.”

There are 10 dots of each colour.  Broadly speaking, Australia and the UK have chosen the government regulation approach.  In Australia, every major political party seems to agree that the “solution” is always more (they would say better) regulation.  In the UK, the Lib Dems and Tories make occasional mutterings suggesting that they might agree with me, but for the most part they’re in lock step with Labour (UK), which likes the status quo.

America is a bit more varied.  By and large, they have adopted the approach of letting tort law and the fear of civil suits induce the effect of (remarkably strict) regulation, but when America does use explicit government regulation, it tends to be something of a light touch.  Democrats seem to want to move closer to the British/Australian model, while Republicans seem to either like their status quo or wish to move to the Libertarian ideal.

Small government / Libertarian idealists typically want no government regulation and, to the extent that things need to be dealt with at all, they want everything to happen through the courts.  Although they want small government, what government they do want, they want to be strong (e.g. in the enforcement of the law).

Speaking about America, M.S. in the above-linked-to Economist entry writes:

I would gladly join any movement that promised to do away with this sort of nonsense. For example, Philip K. Howard’s organisation “Common Good” (TED talk here) works on precisely this agenda. Common Good’s very bugaboo is useless, wasteful legal interference in schools, health care, recreation, and so on. But what you quickly note with many of these issues is that they’re driven by legal liability concerns. You have a snowblader in Colorado suing a resort because she crashed into someone. You have states declining to put up road-hazard signs because the signs prove they knew the hazard was there, which could render them liable for damages. You have the war on children’s playgrounds. The Massachusetts swimming ban, too, is driven by liability concerns. The park officials in Massachusetts aren’t really trying to minimise the risk that you might drown. They’re trying to minimise the risk that you might sue. The problem here, as Mr Howard says, isn’t simply over-regulation as such. It’s a culture of litigiousness and a refusal to accept personal responsibility. When some of the public behave like children, we all get a nanny state.

Which is exactly what I’m saying about America in my summary, but I think (at least, from my reading) that M.S. is assuming that the opposite of a litigious society is personal responsibility.  That’s not true, I’m afraid.  The level of personal responsibility is orthogonal to whether your society chooses litigiousness or state regulation.

Nevertheless, I suspect that M.S. (and Matt Yglesis) and I are on the same side in this debate.  Let people decide for themselves; they’re adults, or should be.  Don’t put a nappy (diaper) on me just because you’re mollycoddling idiot number 8,749 over there.

Posted on

Paying interest on (excess) reserves (Updated)

The U.S. Federal Reserve is currently paying 0.25% interest on the reserve accounts of depository institutions.  This is therefore, at present, the primary rate of policy concern (as opposed to the Fed Funds rate):  if a bank can’t get a rate of return that, when adjusted for risk, is greater than 0.25%, they will stick their money in their reserve account at the Fed.  Among others, Scott Sumner [blog] has called this policy a mistake.

There is an economic cost to the policy.  0.25% isn’t much, but it’s the risk-free aspect that complicates things.  If banks’ risk aversion or their perception of the risks associated with investments are high, then a truely risk-free 0.25% could look quite attractive.  With the interest rates on US treasuries so low, there’s certainly reason to believe that risk aversion is still abnormally high at the moment.  Whether the demand for loans is coming from particularly risky projects, or is perceived to be, I don’t know (is there any way of knowing?).

So why have it at all?  I suppose I support the paying of interest on required reserves.  The banks don’t get a choice with them, so it seems only fair that they be compensated.  But for excess reserves, there would need to be an offsetting benefit to justify the policy.  One benefit will be that the interest is paid with new money, so it’s a way of quietly helping banks improve their balance sheets.  There’s currently about US$1 trillion in excess reserves, so that’s about US$2.5 billion per year.  That may be a lot of money to you and me, but it’s not much more than a rounding error to the US banking system as a whole.  Still, it’s something.  Another benefit, depending on your point of view, is that by attracting all that money into excess reserves, the Fed sterilised the QE they engaged in last year.  If you feel that the sterilised QE has caused lower long-term interest rates and hold that those rates are the ones that most significantly drive the economy and distinctly dislike inflation, then you’d probably judge the affair to have been a success [I include the weasel words because I am no longer certain].  A third benefit, which is really a further justification of the second, is that there is evidence that the Fed’s QE appears to have lowered not just US rates, but foreign rates as well.  In that case, then you probably want to sterilise the fraction going to other countries (bad enough, one might think, that America is fixing the rest of the world; it would be unthinkable if America also had to suffer inflation by doing so).

Anyway, all of that is by way of getting around to this point:  via Bruce Bartlett, I’ve just discovered that Sweden also pays interest on reserve deposits, normally 0.75 percentage points lower than their repo rate.  But, crucially, their repo rate is currently only 0.50%, which means that their deposit rate is negative, at -0.25%.

For myself, I tend to think that the interest rate on excess reserves should be lowered.  My argument is similar to what I imagine Scott Sumner would say, so I should also explain his view a little, to the extent that I understand him.  With nominal GDP at US$14 trillion, the US$1 trillion sitting in excess reserves is a very, very large amount of money.  If it were released into the economy, it would be a huge stimulus (even if the money multiplier/velocity of money is temporarily low).  By choosing to sterilise their QE (presumably out of fear of inflation), the Fed has turned what could have been a tremendously effective stimulus into a mediocre one at best.  Scott is rather more sanguine about inflation in general than I am (he favours targeting NGDP; I suspect that this graph would make him want to tear his hair out), but even if the Fed wishes to target inflation of, say, the near-universally accepted benchmark of 2%, then with actual current inflation down at 0.5% and expected future inflation below 1.5% for most of the next 10 years and falling, the sterilisation has been excessive.

Update 6 Aug 2010:

The FT’s Alphaville has gathered the arguments for and against.  Here are three arguments (and their counter-arguments) for keeping the Interest on Reserves (IoR) unchanged:

First, from Ben Bernanke himself, made in recent congressional testimony:

The rationale for not going all the way to zero has been that we want the short-term money markets like the federal funds market to continue to function in a reasonable way because if rates go to zero there will be no incentive for buying and selling federal funds, overnight money in the banking system, and if that market shuts down … it’ll be more difficult to manage short-term interest rates, for the Federal Reserve to tighten policy sometime in the future. So there’s really a technical reason having to do with market function that motivated the 25 basis points interest on reserves.

I think this is silly. It’ll be more difficult to manage short-term interest rates in the future only if, following an effective shut-down of the federal funds market, it becomes costly to start it back up again. I seriously doubt that the banks are going to take their existing staff, processes and infrastructure dedicated to this and throw them out the window. Heck, in a Q&A session after his testimony, Mr Bernanke stated that lowering the interest rate on reserves is a (serious) option in the event that the FMOC decides that further stimulus is warranted:

But broadly speaking, there are a number of things we could consider and look at; one would be further changes or modifications of our language or our framework describing how we intend to change interest rates over time — giving more information about that, that’s certainly one approach. We could lower the interest rate we pay on reserves, which is currently one-fourth of 1%.

A second viewpoint, put forward by Dave Altig (of the Atlanta Fed) and Joseph Abate (of Barclays Capital), is that

If banks didn’t get interest from the Fed they would shift those funds into short-term, low-risk markets such as the repo, Treasury bill and agency discount note markets, where the funds are readily accessible in case of need. Put another way, Abate doesn’t see this money getting tied up in bank loans or the other activities that would help increase credit, in turn boosting overall economic momentum.

I think that Jim Hamilton’s response to this is excellent, so let me just quote it in full:

But Dave doesn’t quite finish the story. If I as an individual bank decide that a repo or T-bill looks better than zero, and use my excess reserves to buy one of these instruments, I simply instruct the Fed to transfer my deposits to the bank of whoever sold it to me. But now, if that bank does nothing, it would be left with those reserve balances at the end of the day on which it earns nothing, whereas it, too, could instead get some interest by going with repos or T-bills. The reserves never get “shifted into short-term, low-risk markets”– instead, by definition, they are always sitting there, at the end of the day, on the balance sheet of some bank somewhere in the system.

The implicit bottom line in the Abate story is that the yields on repos and T-bills adjust until they, too, look essentially to be zero, so that banks in fact don’t care whether they leave a trillion dollars earning no interest every day.

The essence of this world view is that there are two completely distinct categories of assets– cash-type assets which pay no interest whatever, and risky investments like car loans that banks don’t want to make no matter how much cash they hold.

But I really have trouble thinking in terms of such a two-asset world. I instead see a continuum of assets out there. As a bank, I could keep my funds overnight with the Fed, I could lend them in an overnight repo, I could buy a 1-week Treasury, a 3-month Treasury, a 10-year Treasury, or whatever. Wherever you want to draw a line between available assets and claim those on the left are “cash” and those on the right are “risky”, I’m quite convinced I could give you an example of an asset that is an arbitrarily small epsilon to the right or the left of your line. Viewed this way, I have a hard time understanding how pushing a trillion dollars at the shortest end of the continuum by 25 basis points would have no consequences whatever for the yield on any other assets.

Finally, back with Dave Altig, there is the argument that:

the IOR policy has long been promoted on efficiency grounds. There is this argument for example, from a New York Fed article published just as the IOR policy was introduced:

“… reserve balances are used to make interbank payments; thus, they serve as the final form of settlement for a vast array of transactions. The quantity of reserves needed for payment purposes typically far exceeds the quantity consistent with the central bank’s desired interest rate. As a result, central banks must perform a balancing act, drastically increasing the supply of reserves during the day for payment purposes through the provision of daylight reserves (also called daylight credit) and then shrinking the supply back at the end of the day to be consistent with the desired market interest rate.

“… it is important to understand the tension between the daylight and overnight need for reserves and the potential problems that may arise. One concern is that central banks typically provide daylight reserves by lending directly to banks, which may expose the central bank to substantial credit risk. Such lending may also generate moral hazard problems and exacerbate the too-big-to-fail problem, whereby regulators would be reluctant to close a financially troubled bank.”

Put more simply, one broad justification for an IOR policy is precisely that it induces banks to hold quantities of excess reserves that are large enough to mitigate the need for central banks to extend the credit necessary to keep the payments system running efficiently. And, of course, mitigating those needs also means mitigating the attendant risks.

But, to me, this really sounds like an argument for having higher reserve requirements, not an argument for encouraging excess reserves.  I’m all for paying interest on required reserves and setting the fraction required at whatever level you judge necessary to ensure the operation of the payments system.  But don’t try to shoe-horn that argument into keeping interest payments on excess reserves.

Posted on

Reporting reactions to the news, not the news

XKCD: Public OpinionI know I’m not alone in getting frustrated by the tendency, in all forms of mass media, to report on reactions to an event or debate rather than provide substantial detail on the event or debate.  I do realise that it’s because the drama of people’s reactions keeps the audience’s attention for longer, that most people aren’t actually interested in the finer points, that it bores them.

Jon Stewart lambasts America’s television news providers for providing anything but news, but for me the sharpest sense of frustration comes when I read a newspaper.  I don’t really blame the providers of news for being consumed by the desire to entertain when they have sound, colour and moving pictures at their command.  Well, okay, I do.  But the defence of the newspaper editor is far weaker.  Sure, there are technicolour tits on page three, but other than that and an over-sized font for the headlines, there’s not much the newspaper can do to distract you from the article itself.

Most people don’t read more than the first few paragraphs of an article.  That’s why papers like the NY Times put those delicious, tantalising nuggets on the front page for the vrapid browsers among us and then send the hungrier reader off to page Q13, or whatever, to finish the piece.  It’s not a practice we see in Britain, but I quite like it.  It gives a visual honesty to our collective consumption of news.  It lets me imagine, as I hunt through the paper for section Q, that the real meat of the article, the guts, the nitty gritty, the actual news, is available in there somewhere.  Sadly, it almost never is.

I don’t want to single out The Grey Lady.  There is no paper anywhere on earth that consistently lists out the facts in each article.  I don’t even need quality writing.  Just chop off the final paragraph and replace it with the facts in bullet point form.  Nobody reads that paragraph anyway, even if it is the one the journalist fought most with the editor to keep.  Leave the rest of the article peppered with Mr. and Mrs. Jones’s sob story and some politician’s outrage, but give me the facts quietly at the end, where it’s not hurting anyone.

Anyway, via Matt Yglesis, I see that a report has been written by Pew Research on the coverage of the health care debate in America.  You can see the full report here or a summary here.  I quite agree with Matt that the most telling aspect of the report is summarised in the following graph (although I disagree with his conclusion that this is not such a bad result):

Pew: Top Health Care StorylinesIt’s a terrible diagram, because 3D graphs make it near-impossible to read the actual numbers (I wonder if Pew Research sees any irony in trying to present these data in a snazzy format), so let me give them to you:

  • 41% : Politics and strategy
  • 23% : Descriptions of [proposed] plans
  • 9% : [Current] State of health care
  • 8% : Legislative process
  • 6% : Obama’s health care plan
  • 4% : Town hall protests

This is for all forms of media, though.  The then current state of health care featured more prominantly in newspapers, which gave it 18% of their coverage.  That’s better, but I suspect it’s deceptive.  That 18% will have included innumerable emotion-dripping sob stories about some old lady and her dodgy hip.  Disappointingly, online news sites, which have essentially zero marginal cost for an additional paragraph on the end of a story, gave only 8% of their coverage to describing the then current system.

Ah, well.  Go read the report.

Update:  Ezra Klein makes an excellent point:

It’s trite to say it, but the news business is biased toward, well, news. There are plenty of outlets that tell you what happened yesterday, but virtually no organizations that simply tell you what’s going on. Keeping up on the news is easy, but getting a handle on an ongoing situation that you’ve not really been following is hard. In recent years, we’ve seen the rise of outlets like FactCheck.org, which try and police lies that are relevant to the debate. But there’s really no one out there who is trying to give you the background to everything going in the debate. News organizations will write occasional pieces trying to sum up the legislation, but if you miss them, it’s hard to find them again, and they’re not comprehensive anyway. The fact that I still can’t direct people to one really good, really clear, really comprehensive online summary of the bill is an enduring frustration for me, and a real problem given the importance of the legislation and the number of questions there are about it.

If I edited a major publication — or even a medium-size one — I would begin each major legislative battle by detailing a few of my smartest, clearest writers to create a hyperlinked, fairly comprehensive, summary of the basic legislation. That summary would be updated throughout the process, and it would be linked in every single story written on the topic. As reader questions came in, and points of confusion arose, it would be expanded, so by the end, you’d have a document that was current, comprehensive, navigable and responsive to the questions people actually had about the legislation. Telling people what just happened is undeniably important, but given that most people aren’t following that closely, we in the media need to do a better job of telling people what’s been happening.

Posted on

Lost

The last 20 minutes of the last episode of the last season were the only 20 minutes I watched.

Conclusion: Americans are so f’ing religious.

The last 20 minutes of the last episode of the last season were the only
20 minutes I watched.

Conclusion:  Americans are so f'ing religious.
Posted on

America and health care

In the light of the recent passage by the U.S. House of Represenatives of the Senate’s version of healthcare reform and the ensuing wailing, gnashing of teeth and smearing of soot in the hair by opponents of said reform, let me give my view – as an outsider – on the matter:

It’s a question of morality.

It astounds me — and, frankly, every other non-American USA-watcher in the developed world — that the richest nation on earth, whose very constitution proclaims the pursuit of life, liberty and happiness to be it’s highest ideals, whose citizenry so loudly profess to live by Christian virtues, would not guarantee that some form of basic, minimum healthcare be available to all of its citizens independently of their ability to pay.  It utterly astounds me.  If I were American, it would disgust me that this had not happened 50 years ago.

If my income and my wealth is above average for my society, I have an ethical duty to subsidise the health care of those who are, for whatever reason, at the lower end of the spectrum.  Yes, there are issues of free riders and of personal responsibility, but they simply do not matter when answering the basic question.  The government of a country, acting on behalf of that country’s people, has a moral imperative to provide a minimum level of care to all of its citizens.

I am not saying this as a screaming socialist.  I freaking hate socialism.  I love the market (when it’s allowed to function properly with full transparancy).  I support (at least partially, and possibly fully) privitised social security.  I like the idea of small government.  I rage against the nanny-state in Australia and in the UK.  I worry about encouraging dependency and a sence of entitlement in those people assisted by the government.  But those concerns take a back seat on this issue.

So, yes, the second question (a two-for) is to ask what the minimum level should be and how to pay for it.  But first question should have been a no-brainer.

If all the country can afford is a polio shot and a packet of aspirin, then that’s what they should provide (hopefully a charity or two might help out, too).  But if the country is the richest in the history of the planet, they should be able to stump up for a bit more.

And, yes, for the next criticism, this particular reform by the U.S. Congress is nominally promising more than it will reallly provide when it comes to the fiscal deficit.  Yes, again, given America’s political structure, U.S. government spending won’t be truely corrected until there is a real crisis approaching (as opposed to the make-believe crises being proclaimed by people opposed to the bailouts and stimulus package(s)).

I don’t care.  The child of an unemployed, drug-taking high-school dropout should not be deprived of basic access to a doctor just because we’re angry at their parents.  Nor should their parents, come to that.

Posted on

Party discipline in the Republican Party

Inspired by this post by Cam Riley … Any observer of U.S. politics could not have failed to notice the incredible level of party discipline that the Republicans, particularly in the Senate, have achieved over the last year or six.  This may be something new to Americans, but it’s rather common to Britons and Australians, who generally get more excited when somebody — anybody! — breaks the party line.  The party discipline of the Australian Labor Party, in particular, is phenomenal.

I understand that the generally accepted explanation for the differences between the USA and Australia in this regard focuses on the sources of funding for campaigns.  In Australia, all campaign funds come from the party — individual candidates cannot raise money directly — where as in the US, there’s a combination of party-supplied and individually-raised funding.

That then suggests two possible reasons for the new-found Republican discipline:

  • Republican congressional candidates have started to take a larger fraction of their total campaign funding from the party itself; and/or
  • Advocacy groups that support policies we stereotypically associate with the Democratic Party have not been giving any money to Republicans.

If it is the second reason, then that is a tactical error, and a foolish one, on the part of those advocacy groups.

Posted on

Clues they missed

The NY Times has a nice piece summarising the clues that were not put together prior to the attempted Christmas Day attack on the Northwestern flight to Detroit.  All of those clues are presented in this graphic:

The article speaks of poorly designed computer systems, phrases mentioned in speeches (which, frankly, would be incredibly difficult to automatically include in your analysis) and general stories of the left hand not knowing what the right hand is doing within the US intelligence community.  To my mind, though, it really comes down to just a few simple points:

  • His own father had contacted the US to say that his son was missing and possibly being “radicalised,” leading to his being placed on a watch list.
  • He bought the ticket with cash.
  • He checked no luggage.

I firmly believe that most airport security is theatre, but the combination of just those three points should surely have warranted an individual pat-down.

Posted on

More on the US bank tax

Further to my last post, Greg Mankiw — who is not a man to lightly advocate an increase in taxes on anything, but who understands very well the problems of negative externalities and implicit guarantees — has written a good post on the matter:

One thing we have learned over the past couple years is that Washington is not going to let large financial institutions fail. The bailouts of the past will surely lead people to expect bailouts in the future. Bailouts are a specific type of subsidy–a contingent subsidy, but a subsidy nonetheless.

In the presence of a government subsidy, firms tend to over-expand beyond the point of economic efficiency. In particular, the expectation of a bailout when things go wrong will lead large financial institutions to grow too much and take on too much risk.
[…]
What to do? We could promise never to bail out financial institutions again. Yet nobody would ever believe us. And when the next financial crisis hits, our past promises would not deter us from doing what seemed expedient at the time.

Alternatively, we can offset the effects of the subsidy with a tax. If well written, the new tax law would counteract the effects of the implicit subsidies from expected future bailouts.

My desire for a convex (i.e. increasing marginal rate of) tax derives from the fact that the larger financial institutions are on the receiving end of larger implicit guarantees, even after taking their size into account.

Update:  Megan McArdle writes, entirely sensibly (emphasis mine):

That implicit guarantee is very valuable, and the taxpayer should get something in return. But more important is making sure that the federal government is prepared for the possibility that we may have to make good on those guarantees. If we’re going to levy a special tax on TBTF banks, let it be a stiff one, and let it fund a really sizeable insurance pool that can be tapped in emergencies. Like the FDIC, the existance of such a pool would make runs less likely in the shadow banking system, but it would also protect taxpayers. Otherwise, with our mounting entitlement liabilities, we run the risk of offering guarantees we can’t really make good on.

I agree with the idea, but — unlike Megan — I would allow some of it to be collected directly as a tax now on the basis that the initial drawing-down of the pool came before any of the levies were collected (frustration at the political diversion of TARP funds to pay for the Detroit bailout aside).