Break up the megabanks already (#OWS)
For the past few months at Occupy we’ve been focusing more and more on having a single message and goal. That has been to break up the big banks.
What’s great about this goal is that it’s a non-partisan issue; there is growing consensus (among non-bankers) from the left and the right that the current situation is outrageous and untenable. What’s not great, of course, is that the situation is so easy to spot because it’s so heinous.
Yesterday another voice joined the Break-Up-The-Big-Banks chorus in the form of an editorial at Bloomberg (hat tip Hannah Appel). They wrote a persuasive piece on breaking up the big banks based on simple arithmetic involving bank profits and taxpayer subsidy. Even the title fits that description: “Why Should Taxpayers Give Big Banks $83 Billion a Year?”. Here’s an excerpt from the editorial (emphasis mine):
…Banks have a powerful incentive to get big and unwieldy. The larger they are, the more disastrous their failure would be and the more certain they can be of a government bailout in an emergency. The result is an implicit subsidy: The banks that are potentially the most dangerous can borrow at lower rates, because creditors perceive them as too big to fail.
Lately, economists have tried to pin down exactly how much the subsidy lowers big banks’ borrowing costs. In one relatively thorough effort, two researchers — Kenichi Ueda of the International Monetary Fund and Beatrice Weder di Mauro of the University of Mainz — put the number at about 0.8 percentage point. The discount applies to all their liabilities, including bonds and customer deposits.
Big Difference
Small as it might sound, 0.8 percentage point makes a big difference. Multiplied by the total liabilities of the 10 largest U.S. banks by assets, it amounts to a taxpayer subsidy of $83 billion a year. To put the figure in perspective, it’s tantamount to the government giving the banks about 3 cents of every tax dollar collected.
The top five banks — JPMorgan, Bank of America Corp., Citigroup Inc., Wells Fargo & Co. and Goldman Sachs Group Inc. – – account for $64 billion of the total subsidy, an amount roughly equal to their typical annual profits (see tables for data on individual banks). In other words, the banks occupying the commanding heights of the U.S. financial industry — with almost $9 trillion in assets, more than half the size of the U.S. economy — would just about break even in the absence of corporate welfare. In large part, the profits they report are essentially transfers from taxpayers to their shareholders.
Next time someone tells me I want to take money out of rich people’s pockets (and that makes me a free market hater), I’m going to remind them that every time I pay taxes, 3 cents out of every dollar (that I know of) goes directly to the banks for no good reason whatsoever except the fact that they have the lobbyists to support this system. They’re bullies, and I hate bullies.
So no, I’m not suggesting we take honestly earned money out of the pockets of those who deserve it, I’m suggesting we stop stuffing insiders’ pockets with our money. Big difference.
But it’s not just money I object to – it’s future liability. There’s now an established track record of discovered criminal acts that don’t get anyone at the big banks in trouble. We are setting ourselves up for an even bigger bailout of some form soon, one that we taxpayers really may not be able to afford.
I think of the too-big-to-fail problem as like having an alcoholic brother-in-law who not only sleeps on your couch every night but also knows the PIN code on your ATM card. The money is irksome, no doubt, but what if that guy fell asleep smoking a cigarette and me and my kids die in the resulting fiery inferno? And it’s not that I think all addicts could be magically cured, but I don’t want them to have access to my personal stuff. Get them out of my house.
So can we break up the megabanks already? I’d really like to stop worrying about them because I have better things to do.
mathbabe 
You don’t want any (alcoholic) bullies to have access to your personal stuff. That sounds fair enough.
But if you don’t mind that bureaucrats with tremendous power do take (part of) your personal stuff to redistribute it as they please, how can you be surprised of the inevitable outcome of bully pigs using any means available to get at the front row at the feeding trough? If you think the solution will be in replacing above bureaucrat by another, I’m thinking about Tolkien’s work: does it really matter who is wearing the Ring of Power?
About “breaking up banks”: why not just stop bailing them out? They will break up by themselves 🙂
LikeLike
We tried that with Lehman.
LikeLike
And what exactly was the problem for the 99% with what happened to Lehman? Please explain…
PS. excuses for the avatar name change, apparently I’m messing up something in WordPress using multiple names at multiple sites
LikeLike
As I recall, when Lehman failed, the stock market dropped and then the unemployment rate exploded.
LikeLike
Are you really suggesting that we need banks, and inflated stock prices, for jobs? If so, please explain how?
As I recall, mankind has been able to put its muscle/brain power into productive use far before things like money, banks or a stock market appeared.
LikeLike
Do you not understand what the financial sector does?
As I recall, money systems have existed in all history.
LikeLike
I’m not convinced that we need banks or inflated stock prices for jobs. I am however convinced by previous evidence that if bailouts go away, so do jobs for the 99%. If you want to make an argument that jobs will come back, feel free to do so, but make sure you also include an argument for how the 99% would be provided for in the intervening time.
LikeLike
??
Fannie/Freddie got bailed out. Ford, Chrysler and GM got bailed out. Most large banks got bailed out. AIG got bailed out. Almost _every_ non-performing business (of course, only if it was enough “well-connected” to the bureaucrats-in-power) got bailed out. If one thing is clear, the no 1 highlight in our word cloud of government interventions during the last 5 years is “bailout”.
Please explain how you come to the hypothesis that “if bailouts go away, so do jobs for the 99%”, if evidence shows that the amount of bailouts only increased since 2007. An alternative hypothesis could just as well be that the unemployment is caused by the bailouts themselves.
LikeLike
So CitiBank had only 1/10 of its assests before the bailout. ($30 Bln vs. $300 Bln) Yet Too Big To Fail.
Bad business decisions & poor investments will keep you intact. (And Small.)
The US Gov. gambled that Lehman Bros. was small enough to fail & lost that bet. The ‘Original’ Fiscal Cliff!
You’re an insider, aware of this House of Cards.
Bank lobbying is outdoing public input on new regulations. We need experts on our side. Mobilize the lawyers, etc.
LikeLike
There is a hard limit on the amount of “public input” on regulations: the 99% may cast their vote once in a while.
There is practically no limit to the lobbying power of the bullies: they get a positive ROI on their investment, if not they would not lobby.
Guess who wins in this game?
LikeLike
Great column.
Lots of good stuff written on this but I have a new favorite — “The Bankers’ New Clothes” by Anat Admati and Martin Hellwig which argues that the bankers,like the Emperor in the fairy tale, are naked. They say “The jargon of bankers and banking experts is deliberately impenetrable. This impenetrability helps them confuse policymakers and the public”.
“We want to encourage people to form and to trust their opinions, to ask questions, to express doubts, and to challenge the flawed arguments that pervade the policy debate. It we are to have a healthier financial system, more people must understand the issues and influence policy.”
Then they do a good job penetrating the jargon and informing the debate. The regulators aren’t going to fix this unless we make them.
LikeLike
Why don’t we just relax the regulations on banks instead (and I dont mean limits on reserves to lending – those should be VERY strict), so that instead of needing 100 million dollars and a whole lot of political clout plus 5 years of regulatory procedures, you can just set one up with minimal beaurocratic meddling required ? I’d sure as hell be happy to bank with Bank of Amazon for example. I don’t even want interest on my deposits, I just want functional banking…. how hard can that be to set up ? Not hard at all if it wasnt for the regulations.
Regulating that we split up big banks is just EVEN MORE REGULATION and the problem is already the amount of regulation.
LikeLike
Because a lot of people would invest their money with the bank of Enron and it would disappear.
LikeLike
As opposed to what, the Bank of Greece ? Enrons are relatively few and far between, I think most people would spread the risk a bit – or even better take out “bank” insurance ?
LikeLike
“As opposed to what,…”
What are you talking about?
“I think most people would spread the risk” – citation? People who worked for Enron didn’t spread the risk (and put their retirement plans in Enron stock) w/ no implied (or given) bailouts. The “insurance” on the bank is exactly what the government provides and why banking institutions are regulated.
Do that, and you’re back to square one you either pay for regulation, or you pay for an insurance scheme (and hope that there is no counterparty risk) or you spend a large amount of time searching and “doing diligence.”
TAANSTAAFL
LikeLike
Your argument is so incoherent i’m not sure what your point is. There’s so much regulation on banks that there is limited competition and a tendency for them to merge to even further reduce the competition. Smaller banks occasionally going bust would not be much of a problem as it could easily be handled by either private or public insurance. People putting ALL their investments into ANYTHING are risking EVERYTHING, regardless of whether there is fraud or not.
LikeLike
It seems like you’ve come to the conclusion that the only parameter is the total regulation, and that the quality of the regulatory apparatus does not matter and that no argument will dissuade you from your conclusion.
“There’s so much regulation on banks that there is limited competition and a tendency for them to merge to even further reduce the competition.”
Competition in finance is NOT like competition in other fields. In particular, it is impossible (or difficult) for an outsider who can only see returns to tell the difference between a superior returns derived from taking on a lot of risk (which leads to short-run high returns and long-run annhiliation) and superior returns derived from better arbitrage arbitrage and better management/technology (which leads to short-run high returns and long run high returns). Compare this with say your average manufacturing industry, where competition usually leads to technical innovations and lower prices. Worse still, since our proprietary traders and Randian heroes the payment processes and transactions for all large scale industry, their potential demise endangers all productive industries dependent on contracts with investment/commercial (like farming, tech and manufacturing).
Cathy likes the drunk spending all your money analogy, but I like to imagine a doctor who has the drugs you need to live (commercial banking system in particular MMFs), and a gambling problem (prop trading) and is indebted to the mob (da free market). If the doctor loses too many bets and cannot pay the mob back, the mob will kill him (bankruptcy) and you cannot get the drugs you need. If you don’t keep funding his habit, you will die and so you must either pay through the nose (see: TARP) or rig the bets in his favor (see: current FED policy including buying subprimes, QE infinity, the discount window, holding reserves at the Fed, not suing the banks for crimes against the Fed etc. etc.). There are a lot of ways to think about solutions to this problem. You might ban the Doctor from the casino (Glass-Steagall) or try to make casino less risky overall (Dodd-Frank) or let the doctor in the casino but sue him when he plays a high risk game (CFPB) or steal the doctor’s drugs and hope that a new doctor comes along before you run out (nationalization). What you can’t do, what you must not do, is hope that the doctor values his life more than you value yours and let him take the risk the promise that you won’t bail him out – knowing that if the mob kills him you’ll die too.
Anyway, all this is to say that you don’t want to maximize competition (or returns) in the financial sector over a short-time horizon (where short is relative) and there is strong evidence that more free market systems tend to do this, especially in the presence of deposit insurance, hence you need regulation. You might respond, well just get rid of deposit insurance and let all the banks pull themselves up by their bootstraps or whatever. First, there have been periods in history with no deposit insurance and no regulation – those periods often featured large scale banking panics every couple years. Second, withdrawing deposit insurance and regulation might be long-term efficient now (100 years later with better tech or w/e, I don’t think it’s true but let’s entertain your argument) but the main effects of TBTF are in implicit insurance and bailouts (because you cannot let your doctor get killed by bookies!), not in FDIC’s actual insurance on retail deposits. You might say, get rid of TBTF then – no politician can make a credible commitment to do this since the first bank failure would almost certainly cause a depression (see: letting the mob kill the doctor).
“Smaller banks occasionally going bust would not be much of a problem as it could easily be handled by either private or public insurance.”
Lol. They are handled by insurance now, provided by the government. They could also be handled by insurance by the private sector, in principle. I’m not arguing against small banks.
“People putting ALL their investments into ANYTHING are risking EVERYTHING, regardless of whether there is fraud or not.”
I though your point was that we could buy insurance on the one investment? Most people prior to the 1980s had a bank account, and that was it. With government and regulations they did fine.
Anyway, I guess I agree with you. Onwards to libertopia!
LikeLike
Bloomberg’s (and your) reading of the paper is wrong, and 0.8% (which is a HUGE number, in this context) massively overstates the amount by which taxpayers are subsidizing banks. Even if you accept the methodology that Ueda and di Mauro use – and I can think of a lot of reasons to question it (one credit rating agency – and Fitch, at that – as a proxy? Really?) their model does NOT solve for every aspect of a bank’s funding structure – it solves, specifically and narrowly, for a 5 year bond, and that particular type of instrument is a relatively tiny piece of any normal banks capital structure.
So even if one accepts the model as perfect, which I don’t, Bloomberg (you) are out to lunch if they (you) apply it to the whole of the banking systems liabilities. Banks get a huge portion of their funding from deposits: those are specifically and separately insured, at least up to the $250k per a/c limit – so taxpayer insurance means nothing in that context. A lot of banks other activities are funded via (say) overnight repos – which in turn are collateralized. The cost of the repo explicitly depends on the quality of the collateral – NOT on whether a bank is TBTF or not.
One could go on for a while here, but you get the point, no? Certainly Ueda and di Mauro do, which is why they don’t make anything like the claim that Bloomberg does.
I find myself on the opposite end of pretty much every social argument that you advance here, which is great – that’s why I read the blog. But to reprint a blatantly false interpretation of a paper – even if you’re willing to completely overlook some fairly prominent flaws in its methodology – is lazy at best and irresponsible at worst.
And isn’t that exactly what you accuse bankers of being?
LikeLike
I’m willing to accept that the research may be flawed, but let’s not stop there. What is the actual implicit taxpayer subsidy per year? What is that in terms of cents on a taxpayer’s dollar? Instead of saying “wrong” can you correct the arithmetic? These are good ways of framing the discussion even if their numbers are flawed.
Thanks for reading! Cathy
LikeLike
I suspect that one of the reasons that the original paper narrowed its frame of reference so tightly is that the question as you’ve framed it is an extremely difficult one to answer. If you’re asking about the value of an implicit guarantee, you have to address the probability that market participants ascribed to it ever being exercised, as well as what the dollar value of the guarantee was in the event that it ever did get used.
Those aren’t easy questions to answer, and the authors acknowledge that their proxies are kind of broad…. and if you accept “broad”, then you’re also accepting “imprecise”. And that doesn’t mean “wrong”: when I was in Andrei Schleifer’s (I know – one of your all-time favourite guys) Corporate Finance class at GSAS we’d pull papers apart, and to their credit, the authors of this one acknowledge quite a few of the limitations of their model.
So yes, I do have a number of issues with the paper itself – but my REAL problems are with Bloomberg’s mis-reading of it:
– The paper specifically addresses the benefits of government support to a 5 year bond. Bloomberg applies that to the every liability in the bank.
– The authors provide a range of possible values for their solution, and state that the actual number for any bank will be dependent on (among other things) its starting ratings.
This goes back to the point I made above: how valuable a guarantee is depends in part on how likely you are to have to use it. So if some super-sketchy bank that was always on the brink of insolvency suddenly received tax-payer support, that support would be more valuable to its creditors than the support given to a bank that everyone believed to be impregnable.
This isn’t my model, this is the authors’ – and they provide percentage values that range from really pretty tiny at the high end, to mind-blowingly wide at the weaker end of the scale. What Bloomberg does is take the mid-point of the range, and then apply it to a bunch of banks that held pretty decent ratings going in.
So to summarise Bloomberg’s shortcomings on this one, they basically made up a number to reflect the value of a guarantee that doesn’t have any support in the paper they are purportedly referencing. Then, they applied their made up number to a series of banks that they shouldn’t. Finally, they apply it to the every element of those banks’ funding, when the paper specifies that it applies to one very narrowly defined class of instrument.
I’m more than happy to to through my criticism of the methodological flaws in the paper in more detail off-forum – you can see my email, right? But ultimately, plenty of academic work in this area has flaws and is still useful, and that’s how I’d categorize this work.
What’s utterly freaking useless is the way that Bloomberg treated it. They either misunderstood it completely, and didn’t realise it (so, stupid?) or they did understand it and chose to lie about what it says (mendacious). Which do you prefer?
LikeLike
What do you think of this rebuttal of complaints from the Bloomberg editors from this morning?
LikeLike
At first glance, I don’t think Bloomberg’s rebuttal holds water: it does nothing to address the most salient point, which is that the authors took a study which dealt with a very narrow question, took the results out of context, and applied it to a set of variables that the authors stipulate are inapplicable. At best it’s stupid, on any reading it’s far from “reasonable and … transparent”, and at worst it’s actively misleading.
Having said that, addressing their points in the order they raise them:
– Levine isn’t saying that interest rates are low right now, so 0.8% is an unrealistic number for a subsidy: he’s saying that your deposits are already federally insured, so federally insuring them AGAIN (which is what a taxpayer guarantee would do) isn’t going to make you, the depositor, willing to lend to them at 80bps less than you’d charge a small bank.
– I read Jacewitz & Pogach for the first time about 2 minutes ago, so I’m more than willing to admit to missing some of the nuances of their paper. However, I’d highlight a couple of points that stand out: firstly, that they stipulate that the difference in cost could be due to a number of factors: “It does not allow us to identify what it is about large banks that causes the difference.” Off the top of my head, I can think of global relationship management, clients’ perceived need to have a one-stop-shop for financial services, clients’ unwillingness to constitute a significant percentage of a small bank’s deposit base…. AS WELL AS a de facto taxpayer guarantee. Secondly, of course, the range that J&P come up with under most conditions is substantially less than 80bps….. and the absolute dollar value that they impute to the subsidy is commensurately less.
– The use of Ueda’s model for the sensitivity of a banks funding costs to a perceived change in creditworthiness yields a delta of 0.5% (paragraph 18?). Switching to a “more relevant measure” in the next paragraph – which conveniently yields a 1.2% advantage (not to mention changing the time scale in question) should be something of a red flag to someone as attuned to gaming in models as you are, Cathy.
– I’ll agree that there isn’t a great way to do this analysis: I’m not smart enough to know how to model the kind of feedback loops and causality traps that Bloomberg seems so comfortable with.
Ultimately, I don’t think their rebuttal holds up. They took a model that examined the impact on funding costs for a very narrow set of instruments, and applied it to every liability on the balance sheets of their sample set. To come up with a series of ex poste justifications for their analysis (not to mention the single point solution for “subsidy”) doesn’t alter the fact that they distorted the model they purportedly relied upon, and generated a story that can’t be supported by the paper that it’s supposed to be based on.
LikeLike
Aren’t you also glossing over the distinction between the value of the implicit guarantee to the bank versus its cost to taxpayers? Even if the banks are truly enjoying an $83B/year benefit, it is certainly not equivalent to taxpayers writing a check to the banks for that amount.
LikeLike
Either nationlise ’em or make them accountable, banks can’t have it both ways?
LikeLike