Sunday, August 10, 2014

Anat Admati profile in the New York Times

Source: New York Times
When she talks, banks shudder. A very nice profile of Anat Admati by Binyamin Applebaum in the New York Times.

The article got most of the big points right. Banks don't "hold" capital, they issue it.
“The industry has benefited from, and sometimes encouraged, public confusion. Banks are often described as “holding” capital, and capital is often described as a cushion or a rainy-day fund. “Every dollar of capital is one less dollar working in the economy,” the Financial Services Roundtable, a trade association representing big banks and financial firms, said in 2011. But capital, like debt, is just a kind of funding. It does the same work as borrowed money. The special value of capital is that companies are under no obligation to repay their shareholders, whereas a company that cannot repay its creditors is out of business."
Look for the usage "banks hold capital" in the vast majority of financial press, including newspapers that should know better, for a sense of how pervasive this fallacy is.

The article mentioned the argument that equity costs more than debt, got right that much of that is due to debt subsidies and the difference between private and social cost:

A 2010 analysis funded by the Clearing House Association, a trade group, concluded that an increase of 10 percentage points in capital requirements would raise interest rates by 0.25 to 0.45 percentage points. 
This, in the view of Ms. Admati, is a small price to pay for fewer crises. She notes that debt is cheaper than equity largely because of government subsidies — not just deposit insurance but also tax deductions for interest payments on other kinds of debt — so more equity would basically transfer costs from taxpayers to banks. Even in the short term, she says, the economic impact may well be positive. A study last year by Benjamin H. Cohen, an economist at the Bank for International Settlements, found that banks with more capital tended to make more loans. 
The article couldn't quite get to the Modigliani-Miller point that the cost of equity declines if banks issue more of it. Still, for newspaper coverage of tough issues, this was really good.


I'm a big fan of Anat's courageous crusade for capital. (For example, my review "The banker's new clothes," and "Toward a run free financial system" also arguing for more capital.)  Anat doesn't just sit around and write essays, opeds, and occasional snarky (or "polemical") blog posts. Anat has taken on the hard work of "dogg[ing] from the West Coast to the East Coast to Europe and back again and over again.” She testifies in Congress, she goes to endlessly boring bank regulation conferences, she dukes it out with Vikram Pandit (then Citigroup CEO) on the pages of the FT. When bank apologists write self-serving balderdash, I shrug my shoulders and move on. Anat gets on a plane.

The last sentence:
she said she was glad that policy makers finally seemed to be listening. But, she said, she was frustrated by the lack of progress and not sure about how to press ahead.
To the contrary, this has been one of the most successful campaigns to change ideas in economic policy, in a short time, that I have ever seen. If you want to study how an academic economist can have a major influence on public policy, this is it.  My capsule history (from "Challenges for cost-benefit analysis in financial regulation")
In the Dodd-Frank act, higher capital requirements are a small element in a sea of regulation. But in the subsequent policy discussion, simple and very high capital requirements have come to the fore as probably the best idea that has a realistic chance of success.

As a concrete example, the French et al. (2010) Squam Lake Report written by a team of academic financial economists (including myself) included a short chapter on “reforming capital requirements.” It includes a speculative list of “costs” of capital requirements, including management “discipline” by the threat of a run, and potential “economies of scale.”... 
But this isn’t really the focus of the book’s recommendations to prevent financial crises. Chapters on “systemic regulator,” “new information infrastructure,” “regulation of executive compensation,” “improving resolution options,” two chapters on derivatives and prime brokers, and a clever proposal for “regulatory hybrid securities” really draw the author’s passions.
 In the following years, my own thinking, and I think that of many economists and agencies especially including the Fed, shifted... The larger consensus has shifted away from clever schemes for convertible debt, farsighted benevolent regulators, and any faith in resolution, to capital, just more capital.

Admati and Hellwig (2013) ... argue straightforwardly for more simple equity capital... 
And now, much higher simple capital ratios are the only component of Dodd-Frank that most observers put much faith in. Where 5% was once radical, the idea that 20%, 30% or more capital has very little social cost is now commonplace.
Anat's dogged persistence is a big part of this story.

Interestingly, it is, I think, the simple unimpeachable logic of her position that is carrying the day. Usually policy debates are fought out with complex "studies" with tables of numbers that nobody really understands, and "theory" is disparaged. And plenty of "studies" with big numbers have been written opposing her.

An interesting personal note:
Admati decided to enter the public square because she felt that academics and policy makers weren’t listening. ... She was not sure how to reach a popular audience, so in 2010 she enrolled in a program [] that teaches prominent women to write opinion articles. 
Writing is hard, communicating is hard, and investing in that skill is worthwhile and a worthy example.


  1. I'm curious: Is the ultimate point of increasing capital "holdings" that the banks will be more liquidity to handle negative shocks to the economy (thus preventing crises like in 2008) or that banks will take less risk with more "skin in the game," or both?

    1. I'm not going to do it all in a comment response. Go look at the links. Briefly, capital is not "held," don't even say that. Banks GET money by issuing equity capital or by borrowing. If they get money by issuing equity, when they lose money their stock price goes down. Period. Having made no fixed promises, they cannot go bankrupt = fail to pay their debts.

    2. I think you missed the part where I used quotations to follow the same standard you did, indicating I understand what you mean and am using it as a - I thought - mutually intelligible shortcut.

      I will check out the other links, but briefly, my question is really getting to the transmission mechanisms that Admati and your suggestions work through to make the system less susceptible to crisis. If those don't satisfy, I will come back here.

    3. Professor Admati seemed overly dismissive of the argument that regulation in one area moves activity to another area. Her pollution analogy implies agreement, I would think, unless she is arguing that regulation tends to creep due to shared long term interests.

    4. The article -- very condensed -- gives that impression. Read the book, it's much more detailed. Once you understand MM, you realize that being allowed to go on with little capital is not a cost advantage, it is a subsidy.

    5. Green,

      There is a simple solution to the “moves activity to other areas” problem: regulate the “other areas”. Indeed, it’s totally absurd to impose rules on regular banks while not imposing the same rules on shadow banks, given that the shadow bank industry is almost as large as the regular industry. As Adair Turner (former head of the UK’s Financial Services Authority) put it, “If it looks like a bank and quacks like a bank” it should be regulated like a bank.

    6. Ralph,

      You fail to make the distinction between banks and shadow banks whereby the latter can only impose its losses on private capital contributors, and not the government in any conceivable way.

    7. Anonymous - a run in the shadow banking sector can have such large economic consequences that the government has to step in and buy the shadow banking commercial paper even though there is no legal obligation on the government to do so. The Commercial Paper Funding Program is an example. The government suffered no losses but certainly took on significant risks.

  2. I'm curious of her position on the "right" equity/debt mix for household purchases of assets, namely housing. If banks should capitalize with >30% equity, should that then apply to all actors within the economy? Or is there some sort of optimal "leverage discrimination" which balances "fairness," wealth protection, and economic growth?

    1. Easy. Households are not "systemically important." Their potential bankruptcy does not occasion a run on their debt. They borrow long term. They don't get government bailouts and debt guarantees. Well, they used not to.

    2. Is the federal government "systemically important"? If so, should it rely on equity instead of debt financing?

    3. This comment has been removed by the author.

    4. Modest Merlin,

      That, however, is not how it has worked out. In 2011, the six agencies (the Federal Reserve, the Comptroller of the Currency, the FDIC, the SEC, the Federal Housing Finance Agency, and the Department of Housing and Urban Development) published an earlier version of the QRM. This drew sharp opposition from the mortgage industry, primarily because it included a 20 percent downpayment. Many of the opponents complained that this would make it difficult or impossible for low-income, minority or first-time homebuyers to buy homes.

      Faced with this opposition, and probably pressed by the Obama administration to meet the objections of community activists and others who oppose meaningful mortgage underwriting standards, the agencies seem to have thrown in the towel, deciding that there would be no underwriting standards at all.

    5. "Is the federal government "systemically important"? If so, should it rely on equity instead of debt financing?"

      Since there is no federal government to bail out the federal government, there is no implicit guarantee on federal government debt.

    6. Are we sure that householders aren't systemically important? Sure, in general defaults by homeowners are not correlated, so why would we care if a few go bankrupt. But if 30% of homeowners all go broke at the same time (maybe due to a bubble), that'd be pretty systemic?

    7. Anonymous,

      The guarantee comes from the 14th Amendment to the Constitution. A good article on it can be found here:

      This has the ability to create a conflict between monetary policy and fiscal policy. Monetary policy wants to set an interest rate, but heavily indebted government does not collect enough tax revenue to make the interest payments on the debt.

      The question is - should the central bank bail out the federal government?

    8. Well, if you look at developing countries with high inflation history, they cannot issue long term debt in their currencies. (Not because its forbidden but because noone will buy it)
      So whoever buys debt takes the risk of inflation with it I guess. Thats an important topic but not directly related to implicit government guarantees on private entities debt.

    9. Anonymous,

      "So whoever buys debt takes the risk of inflation with it I guess."

      Domestic buyers take the risk of inflation, global buyers take the risk of currency depreciation - not exactly the same thing. Both buyers are implicitly protected (on a nominal basis) by the U. S. Constitution.

      "That's an important topic but not directly related to implicit government guarantees on private entities debt."

      It becomes directly related when private debt is converted to public debt - see:

      Go to page 19 of 167. In 2008 there was 6.3 Trillion of federal debt and 17.1 Trillion of financial sector debt. Now there is 12.3 Trillion of federal debt and 13.9 Trillion of financial sector debt.

  3. I am convinced that more capital is the right idea. Where is the model?

    1. The model is this - a company that is unable to pay back it's creditors has a few options:,_Title_11,_United_States_Code

      Chapter 7 - Liquidation
      Chapter 11 - Reorganization
      Or the federal government can come along and absorb the debt (private debt becomes public debt - aka bailouts).

      The question is what should be done about a company that is "systemically important". Meaning, the failure of that company could likely result in the failure of a multitude of other companies.

      Neither chapter 7 or government bailouts are good ideas in this circumstance which leaves you with chapter 11 bankruptcy. Under chapter 11, many times a company's debt is converted to equity.

  4. Can't one consider the role of Fannie/Freddie as a debt guarantee mechanism analogous to FDIC insurance? Both subsidize borrowing, albeit of different duration debt, at the cost of the taxpayer at large. Taxpayers presumably benefit but are most-likely not receiving enough premium for the subsidy they are providing. This is of course without including other homeowner non-g'tee subsidies like the mortgage interest deduction, home-sale capital gains exemption, or even Fed MBS QE. It seems likely that some form of the Johnson–Crapo bill will likely reduce housing subsidies sometime in the near future; that is if we don't have another asset-deflation crisis first.

    I'm not familiar with the technical definition of "systemically important," but as you remarked, it seems as though anything worthy of a government bailout is considered pretty important in the eyes of policy makers.

  5. I think the biggest problem with banking is the vague, ever-expanding list of regulations within Dodd-Frank, and the never-ending populist lawsuits from every jurisdiction. So much for simple rules for a complex world, which would be 1) higher capital requirements and 2) a tax on assets above, say, $500B, to discourage large banks that are politically impossible to let fail.

    1. Agree, but who will lobby for that?

    2. Eric Falkenstein,

      I agree that Dodd-Frank is far too complicated, and same goes for the UK equivalent: the so called “Vickers” set of rules, which I’ve looked at in detail. Plus you are right to say that much higher capital requirements solve the problem, and that that is beautifully simply by comparison. Personally I back full reserve banking, that is a 100% capital requirement (as advocated by Milton Friedman, Laurence Kotlikoff and others).

  6. I am thrilled to learn that somebody other than me is concerned with the capitalization of the country's financial companies. Ever since 2008, I have thought that the problem that most needs to be attacked is not regulation, but capitalization.

    As a start the idea of "risk free assets" should be abandoned. It is the central conceit of the Basel "risk based" capital requirements. The one thing we have learned in the last 6 years is that all assets are risky, perhaps in inverse proportion to to their appearance of riskiness. Mortgages were supposed to be safe. Government bonds were supposed to be safe. Ha. Ha. Ha.

    Having dispensed with that bit of nonsense we should set high minimum capital requirements for all banks. Something on the order of $1 for each $8 of assets.

    Further, to induce ultra large institutions to right size themselves, I would add an extra capital requirement to institutions larger than $72 Billion in assets (the top 35). The requirement would be extra percentage points equal to:

    The natural logarithm of its total assets – 25.

    So JP Morgan, which has ~$2.5 Trillion in assets would have an additional capital requirement of 28.5 – 25 = 3.5% or about 16% over all. They might decide to reconfigure the business a bit.

    Further the 35 largest banks would be required to have an equal amount of their funding derive from long term subordinated debentures. Those instruments (~$400 billion for JPM-Chase) would attract a CDS market that would be a canary in the coal mine for them.

    Another issue of capitalization is one that I have raised repeatedly here. The Federal Reserve System has ~4.4T$ of assets and a capital of 56G$, about 1.2%. If the Federal Reserve were a bank regulated by the Federal Reserve, the Federal Reserve would threaten them with receivership, and tell them to shrink their assets or sell a lot more stock.

  7. I always say, KISS, or Keep It Simple Stupid. This is doubly important in government tax codes and regulatory policies.

    Complexity, by tis nature, is undemocratic, and usually mystifying. Behind the cloaks of complexity, a thousand sins or omissions can occur.

    So I like the simple capital solution.

    Okay, but should all banks by regulated to have 100 percent equity? Or just banks that want FDIC insurance?

    If I say. "I am not a bank," but I borrow money at 4 percent and lend it out at 7 percent, what then? Suppose I end up with trillions on both sides of the ledger? Still, not a bank? What is a bank?

  8. It seems that we assume a bank's investment decisions are going to be unrelated to the way it is capitalized. I'm not sure that's a good assumption. If you regulate a more conservative capital structure you could find bank management opting for a more risky, and higher return, investment strategy as a means of offsetting the seemingly lower regulated returns.

  9. So how effectively would higher capital requirements for banks have curbed the warehouse lending that contributed to the proliferation of subprime lenders prior to the crash? Would it have prevented Bear Sterns from issuing commercial paper to finance large purchases of mortgage-backed securities backed by these loans? Now, I am not necessarily against raising capital requirements for banks, but like others have pointed out, I am concerned that financial institutions will simply circumvent the requirements by expanding their shadow banking activities. And this may be even more dangerous.

    1. Increased capital requirements would not prevent a bank from running itself into the ground. It would limit the government's involvement in dissolving the company.

      Increased capital requirements will not solve too big to fail. Even if a company is funded entirely with equity, it can still make bad business decisions.

    2. Right. So as we massively stiffen bank capital, we also have to say that nobody gets to issue large amounts of short term commercial paper or similar run-prone securities in order to finance, say, mortgage-backed securities. That takes some regulation, yes, but orders of magnitude less than the Dodd-Frank army. See "toward a run free financial system" on my webpage for details of my answer to these questions.

    3. OK, I agree. Thanks for the response!

    4. John,

      You are missing something very important here - money is fungible. Company does the following:

      1. Obtains $5 million is short term funding through commercial paper
      2. Obtains $5 million in funding through sale of equity
      3. Makes $5 million in long term loans
      4. Builds a new building for $5 million

      Prove that the money borrowed short term is the same money that was lent long term.

  10. Do you know why Charles Calomiris is critical of Admati's drive for more common equity at banks?

    And if the banking system (regular and shadow) produces "money," won't the drive to make a system less run-prone also reduce the money supply?

    1. Anonymous,

      I presume you are referring to this:

      "On the other hand, the government often finds itself in a somewhat equivocal position when it comes to banks. For it is itself a big borrower and always in need of financing. Where better to turn to than the banks? In many countries, banks were originally licensed during the 17th and 18th centuries precisely so that financially pinched governments could raise money — the two most notable instances being the Bank of England and the Banque de France. To this day banks remain not only a tempting source of cheap funding for governments but also, increasingly, vehicles for channeling politically motivated loans at subsidized rates to important constituents and special interests. There is ultimately no way of getting politics out of banking."

      The way you get politics out of banking is by getting government out of the borrowing business - that means one of two things:

      1. Balanced budget for good times and bad
      2. Government sells equity

    2. There are ways to make the system less run-prone while *increasing* the supply of money, but not everyone agrees that this is necessary or that there is a potential shortage. Prof. Cochrane has a proposal for monetary reform, which he believes will produce an adequate supply of money. His way to protect the system is by making individual banks immune to runs, which clearly prevents a run on the entire system.

      Admati directly addresses your concern about the supply of money in her "Flawed Claims" document which she updates occasionally and posts frequently. You can find her comments easily by searching the document for "money". I think that she points out that less leverage results in greater money-like properties for bank stock. Cochrane has made similar observations on this blog: now that we have continuous price-discovery via high-frequency trading, we might be able to use bank stock for retail payments, especially if banks are not highly-leveraged. He has a paper from 2002 on "Stocks as Money: Convenience Yield ..."

  11. John,

    On banks "holding capital" see this speech from Alan Greenspan (1998):

    "Nor should we require individual banks to hold capital in amounts sufficient to fully protect against those rare systemic events which, in any event, may render standard probability evaluation moot. The management of systemic risk is properly the job of the central banks. Individual banks should not be required to hold capital against the possibility of overall financial breakdown. Indeed, central banks, by their existence, appropriately offer a form of catastrophe insurance to banks against such events."

    What Greenspan is saying is that in the event of overall financial breakdown, the central bank exists as lender of last resort.

    What Greenspan does not mention is that not all financial institutions have access to the Fed's discount window, and so the central bank may not be able to ameliorate a financial crisis through discount window lending if the crisis exists beyond the reaches of central bank influence (shadow banking).

    Also, a systemic crisis may have fiscal origins where solving it puts the central bank's independence at risk.

  12. "In the following years, my own thinking, and I think that of many economists and agencies especially including the Fed, shifted... The larger consensus has shifted away from clever schemes for convertible debt, farsighted benevolent regulators, and any faith in resolution, to capital, just more capital."

    True, but sad. Academics and bureaucrats have given up, while bankers are discredited. The dialogue has broken down and we are now seeing certain unintended consequences.

    I would offer five observations:
    1. It's best to manage risk, not capitalise its consequences. Driving lessons, the Highway Code and traffic cops are better than just plain old "lots more airbags" or titanium crash helmets. By all means, let's have lots of capital, but we can't ignore the role of good risk management, no matter how difficult that might be to a tidy mind.
    2. We need a home for cash deposits. Certain creditors view their cash as "on deposit" rather than an investment in a bank. Either it's guaranteed by the state or it's volatile and flighty. Unfortunately, many MM-type theories ignore this behavioural constraint. Banks -- unlike corporations -- can't be 100% equity financed. If they were, deposits would end up in money market funds, which are quasi-banks and systemically unstable.
    3. New developments -- such as ensuring that reg-cap instruments are truly loss absorbing -- are getting over the massive implicit subsidy of the banking sector and ensuring that there's more "capital at risk" in the system. So long as this capital is permanent ("bail in" of senior short-term debt doesn't work, I'm afraid), then this is progress.
    4. The growth in shadow banking is worrying. P2P and similar new phenomena are prone to problems that will have systemic consequences. We cannot simply allow the leveraged side of banking (whereby deposit-like funds are put to use in loans) to migrate outside the regulated industry -- though this is happening today.
    5. OVERALL, we need a decent amount of leverage in the banking system and really good risk management. "More capital" as a mantra is simplistic and reckless.

  13. "Look for the usage 'banks hold capital' in the vast majority of financial press, including newspapers that should know better, for a sense of how pervasive this fallacy is."
    It's not just the financial press...

    "In all seriousness, though, there’s been a great deal of attention paid to regulatory capital recently, including new Dodd-Frank requirements, Basel III implementation (or non-implementation) issues, and even bipartisan Congressional efforts to raise capital requirements for large banks.[1] Almost all of that attention has naturally centered on the question of how much capital a financial institution should be required to hold." -- former SEC commissioner Daniel M. Gallagher (Jan. 15, 2014)

  14. I am really curious about her. It seems that when she talks, everyone stop to listen and pay attention to what she is saying.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.