Capital can’t be measured

Simon Johnson and James Kwak are absolutely right. Sure, “hard” capital and solvency constraints for big banks are better than mealy-mouthed technocratic flexibility. But absent much deeper reforms, totemic leverage restrictions will not meaningfully constrain bank behavior. Bank capital cannot be measured. Think about that until you really get it. “Large complex financial institutions” report leverage ratios and “tier one” capital and all kinds of aromatic stuff. But those numbers are meaningless. For any large complex financial institution levered at the House-proposed limit of 15×, a reasonable confidence interval surrounding its estimate of bank capital would be greater than 100% of the reported value. In English, we cannot distinguish “well capitalized” from insolvent banks, even in good times, and regardless of their formal statements.

Lehman is a case-in-point. On September 10, 2008, Lehman reported 11% “tier one” capital and very conservative “net leverage“. On September 25 15, 2008, Lehman declared bankruptcy. Despite reported shareholder’s equity of $28.4B just prior to the bankruptcy, the net worth of the holding company in liquidation is estimated to be anywhere from negative $20B to $130B, implying a swing in value of between $50B and $160B. That is shocking. For an industrial firm, one expects liquidation value to be much less than “going concern” value, because fixed capital intended for a particular production process cannot easily be repurposed and has to be taken apart and sold for scrap. But the assets of a financial holding company are business units and financial positions, which can be sold if they are have value. Yes, liquidation hits intangible “franchise” value and reputation, but those assets are mostly excluded from bank balance sheets, and they are certainly excluded from “tier one” capital calculations. The orderly liquidation of a well-capitalized financial holding company ought to yield something close to tangible net worth, which for Lehman would have been about $24B.

So Lehman misreported its net worth, right? Not according to the law. From the Valukas Report, Section III.A.2: Valuation — Executive Summary:

The Examiner did not find sufficient evidence to support a colorable claim for breach of fiduciary duty in connection with any of Lehman’s valuations. In particular, in the third quarter of 2008 there is evidence that certain executives felt pressure to not take all of the write‐downs on real estate positions that they determined were appropriate; there is some evidence that the pressure actually resulted in unreasonable marks. But, as the evidence is in conflict, the Examiner determines that there is insufficient evidence to support a colorable claim that Lehman’s senior management imposed arbitrary limits on write‐downs of real estate positions during that quarter.

In other words, the definitive legal account of the Lehman bankruptcy has concluded that while executives may have shaded things a bit, from the perspective of what is actionable within the law, Lehman’s valuations were legally indistinguishable from accurate. Yet, the estimate of net worth computed from these valuations turned out to be off by 200% or more.

Advocates of the devil and Dick Fuld will demur here. Yes, Lehman’s “event of default” meant many derivatives contracts were terminated prematurely and collateral on those contracts was extracted from the firm. But closing a marked-to-market derivatives position does not affect a firm’s net worth, only its exposure. There may be short-term changes in reportable net worth as derivatives accounted as hedges and not marked-to-market are closed, but if the positions were in fact hedges, unreported gains on other not-marked-to-market assets should eventually offset those charges. Again, the long term change in firm net worth should be zero. There are transaction costs associated with managing a liquidation, but those would be minimal relative to the scale of these losses. Markets did very poorly after Lehman’s bankruptcy, but contrary to popular belief, Lehman was never forced into “fire sales” of its assets. It was and remains in orderly liquidation. Last July, more than 9 months after the bank fell, Lehman’s liquidator reported that only a “fraction” of the firm’s assets had been sold and the process would last at least two years. Perhaps the pessimistic estimates of Lehman’s value were made during last year’s nadir in asset prices, and Lehman’s claimed net worth looks more reasonable now that many assets have recovered. But if Lehman’s assets were so profoundly affected by last Spring’s turmoil that an accurate September capitalization of $28B shifted into the red by tens of billions of dollars, how is it plausible that Lehman’s competitors took much more modest hits during that period? Unless the sensitivity of Lehman’s assets to last year’s markets was much, much higher than all of its peers, Lehman’s assets were misvalued before the asset price collapse, or its competitors assets were misvalued during the collapse.

We get lost in details and petty arguments. The bottom line is simple. The capital positions reported by “large complex financial institutions” are so difficult to compute that the confidence interval surrounding those estimates is greater than 100% even for a bank “conservatively” levered at 11× tier one capital.

Errors in reported capital are almost guaranteed to be overstatements. Complex, highly leveraged financial firms are different from other kinds of firm in that optimistically shading asset values enhances long-term firm value. Yes, managers of all sorts of firms manage earnings and valuations to flatter themselves and maximize performance-based compensation. And short-term shareholders of any firm enjoy optimistic misstatements coincident with their planned sales. But long-term shareholders of nonfinancial firms prefer conservative accounts, because in the event of a liquidity crunch, firms must rely upon external funders who will independently examine the books. The cost to shareholders of failing to raise liquidity when bills come due is very high. There is, in the lingo, an “asymmetric loss function”. Long-term shareholders are better off with accounts that understate strength, because conservative accounting reduces the likelihood that shareholder wealth will be expropriated by usurious liquidity providers or a bankruptcy, and conservative accounts do not impair the real earnings stream that will be generated by nonfinancial operations.

This logic inverts for complex financials. Financial firms raise and generate liquidity routinely. Many of their assets are suitable as collateral in repo markets. Large commercial banks borrow freely in the Federal Funds market and satisfy liquidity demands in part simply by issuing deposits that are not immediately withdrawn. For large financial firms, access to liquidity is rarely contingent upon a detailed audit by a potential liquidity provider. Instead, access to liquidity, and the ability to continue as an operating firm, is contingent upon the “confidence” of peer firms and of regulators. Further, the earnings of a financial firm derive from the spread between its funding cost and asset yields. Funding costs are a function of market confidence, so the value of a financial firm’s real future earnings increases with optimistic valuation. For a long-term shareholder of a large financial, optimistically shading the firm’s position increases both the earnings of the firm and the “option value” of the firm in difficult times. It would be a massive failure of corporate governance if Jamie Dimon or Lloyd Blankfein did not fib a little to make their firms’ books seem a bit better than perhaps they are, within legal and regulatory tolerances.

So, for large complex financials, capital cannot be measured precisely enough to distinguish conservatively solvent from insolvent banks, and capital positions are always optimistically padded. Given these facts, and I think they are facts, even “hard” capital and leverage restraints are unlikely to prevent misbehavior. Can anything be done about this? Are we doomed to some post-modern quantum mechanical nightmare wherein “Schrödinger’s Banks” are simultaneously alive and dead until some politically-shaped measurement by a regulator forces a collapse of the superposition of states into hunky-doriness?

Yes, we are doomed, unless and until we simplify the structure of the banks. When I say stuff like “confidence intervals surrounding measures of bank capital are greater than 100%”, what does that even mean? Capital does not exist in the world. It is not accessible to the senses. When we claim a bank or any other firm has so much “capital” we are modeling its assets and liabilities and contingent positions and coming up with a number. Unfortunately, there is not one uniquely “true” model of bank capital. Even hewing to GAAP and all regulatory requirements, thousands of estimates and arbitrary choices must be made to compute the capital position of a modern bank. There is a broad, multidimensional “space” of defensible models by which capital might be computed. When we “measure” capital, we select a model and then compute. If we were to randomly select among potential models (even weighted by regulatory acceptability, so that a compliant model is much more likely than an iffy one), we would generate a probability distribution of capital values. That distribution would be very broad, so that for large, complex banks negative values would be moderately probable, as would the highly positive values that actually get reported. If we want to make capital measurable in any practical sense, we have to dramatically narrow the range of models, so that all compliant models produce values tightly clumped around the number we’ll call capital. But every customized derivative, nontraded asset, or unusual liability in a bank’s capital structure requires modeling. The interaction between a bank holding company and its subsidiaries requires multiple modeling choices, especially when those subsidiaries have crossholdings. A wide variety of contingent liabilities — of holding companies directly, of subsidiaries, of affiliated or spun-off entities like SIVs and securitizations — all require modeling choices. Given the heterogeneity of real-world arrangements, no “one-size-fits-all” model can be legislated or regulated to ensure a consistent capital measure. We cannot have both free-form, “innovative” banks and meaningful measures of regulatory capital. If we want to base a regulatory scheme on formal capital measures, we’ll need to circumscribe the structure and composition of banks so that they can only carry positions and relationships for which we have standard regulatory models. “Banks’ internal risk models” or “internal valuations of Level 3 assets” don’t cut it. They are gateways to regulatory postmodernism.

Regulation by formal capital has a proud and reasonably successful history, but has been rendered obsolete by the complexity of modern financial institutions. The assets and liabilities of a traditional commercial bank had straightforward, widely acceptable book values. For the corner bank, discretionary modeling mattered only in setting credit loss reserves, and the range of estimates that bank officers, external auditors, and regulators would produce for those reserves was usually pretty narrow (except when all three colluded to fake and forbear in a general crisis). But model complexity overwhelms and destroys regulatory capital as a useful measure for large complex financial institutions. We need either to resimplify banks to make them amenable to the traditional approach, or come up with other approaches more capable of reigning in the brave new world of banking.


Some sources: Yves Smith has been phenomenal on the Lehman bankruptcy. Regarding estimates of the hole that appeared in Lehman’s balance sheet, see “$75 Billion Needlessly Lost in Hasty Lehman Bankruptcy Filing?” and “So Where, Exactly, Did Lehman’s $130 Billion Go?“. Neither Yves nor I remotely buy the “hasty bankruptcy” explanation, see my comments above, the previously cited article, and the always acerbic Independent Accountant. Comments on the pacing of the Lehman liquidation are from the CNBC video embedded in Yves’ piece. The $24B estimated tangible for Lehman is computed by taking its September 10, 2008 shareholder equity and subtracting intangible assets reported on Lehman’s last available balance sheet.

My discussion of financial firms’ incentive to lie was informed by the investor/blogger/super-cop John Hempton, see “Don’t believe what they say” and “Bank solvency and the ‘Geithner Plan’“.

This essay also owes something to Frank Partnoy’s excellent “Make Markets Be Markets” presentation.

Update: Somehow I managed to get the date of Lehman’s bankruptcy wrong. Takes talent, I know. Thanks to commenter mindbender for setting me straight!

 
 

58 Responses to “Capital can’t be measured”

  1. Measuring capital as a stock (as in value) is highly problematic. For this reason, researchers should instead focus instead on measuring capital flows. The perspective of capital flows as the proper measure of capital intentsity was originally argued by Prof Frank Knight back in 1921. Follow the link for a more complete discussion.

    http://wjmc.blogspot.com/2010/01/of-capital-flows.html

    Thank you for the opportunity to comment…

  2. Greg Taylor writes:

    I’ve always thought that financial statements would be much more useful if “margins of error” were attached to the numerical estimates presented. Such transparency might provide a strong managerial incentive to reduce the complexity and riskiness of the firm. As you suggest, margins of error for key measures (i.e capital) would vary widely between firm types and probably within firms of the same type. Investing in a firm that presented a tangible equity interval estimate of -10B and +50B might be viewed as less risky than one with an interval estimate of -100B to +140B, even though each could present a point estimate of 20B.

    I’d agree that it’s really impossible to evaluate a company or rationally invest without some notion of the variability inherent in the financial statements. I’ve never understood why accountants have shirked the responsibility for providing this information. Most accountants seem to think that this is the job of the financial analysts. The accountants seem to me to be much better positioned to provide margins of error since they are producing the estimates.

  3. To Greg Taylor, I completely agree with your comment that variability information should accompany risk estimates. A number is not useful without understanding its context. Financial managers and investors alike must become more polyvalent in their reading of forecasts and risk assessments, which means requiring that the assumptions accompany the estimates.

    http://wjmc.blogspot.com/2010/03/financial-management-requires.html

    Thanks for the opportunity to comment…

  4. glory writes:

    this kinda goes along with MM’s recent missive (and against krugman’s ;)

    What is the market value of a CDS? … The wrong answer is: the market value of a CDS is the chance that the credit default will happen, multiplied by the face value of the CDS. X times Y.

    The right answer is: the market value of a CDS is the chance that the credit default will happen, multiplied by the chance that the CDS issuer will not default, multiplied by the face value of the CDS. X times Q times Y. So, if you buy a CDS on an IBM bond from AIG, its market value is the face value, times the probability that IBM will go bust, times the probability that, if IBM goes bust, AIG will not go bust.

    Now, the people who invented these things are not actually stupid. They didn’t forget about Q. What they said was: Q is the probability that AIG will go bust. They then said: AIG has an AAA credit rating… But this is just bad probability, which makes it bad accounting. The probability that AIG will go bust if IBM goes bust is a conditional probability…

    The basic problem with CDS as a free-market instrument is that, while there is quite a bit of demand for the product of exposure X * Y, there is little or no demand for X * Q * Y… In reality, no party (except USG) is really so much more financially secure than IBM, that it’s worth writing insurance on IBM. The financial product that the market demands is not CDS, but CDS without counterparty risk. This is a useful product indeed, but not a free-market product. No free market can produce any such thing – it takes a fiat-currency issuer…

    Thus, people buy CDS because (and only because) incorrect accounting principles allow them to incorrectly disregard Q, the probability of the CDS issuer defaulting, thus assigning these securities excessive value on their balance sheets. When they become insolvent as a result of this miscalculation, USG’s infinitely-flexible rubber-dollar absorbs the slack, and the Fed becomes the new team sponsor of Manchester United.

    who, btw, just lost to chelsea (ex-rooney!) better luck against bayern :P

    cheers!

  5. Kid Dynamite writes:

    steve, you wrote: “In other words, the definitive legal account of the Lehman bankruptcy has concluded that while executives may have shaded things a bit, from the perspective of what is actionable within the law, Lehman’s valuations were legally indistinguishable from accurate. ”

    but, in fact, wasn’t the conclusion what Lehman’s valuations were LEGAL – and most certainly NOT that they were accurate. It seems to be empirical evidence that the valuations were NOT accurate.

  6. […] the discretion of regulators. Steve Randy Waldman says it’s essentially all irrelevant since we can’t accurately measure capital at large, diverse financial firms. For example: On September 10, 2008, Lehman reported 11% “tier […]

  7. David Merkel writes:

    Hi Steve,

    I think that dumb regulation in concert with a reform of RBC could work to reform the banks. The life insurers do real stress testing, and capital tends to be redundant.

    http://alephblog.com/2010/03/31/a-summary-of-what-bank-reform-should-be/

    http://alephblog.com/2010/03/16/dumb-regulation-is-good-regulation-how-to-regulate-the-banks/

    But one more dumb addition to the rules is needed, and stems from the things that I have written about rating agencies vs. regulators, and level 3 assets. Just as I blame the regulators for outsourcing credit analysis to the rating agencies, particularly for asset classes that are new and have never failed, and thus the loss characteristics are unknown, so I blame the regulators for allowing level three (as measured by GAAP) assets to be measured at any value.

    If we can’t value it neutrally, it should not be a permissible investment, or it should be valued at zero for regulatory purposes. (Like the life insurance industry does with nonadmitted assets.) What cannot be valued cannot be managed. For financials, this is crucial, because financials are a bag of accruals on both sides of the balance sheet, which makes the residual, called capital, many times more variable than the assets or liabilities separately measured.

  8. Maybe we should have two types of banks:

    1. Government run banks that take care of people’s deposits, and have 100% reserves
    2. Commercial banks that give loans, make investments and all the other jazz

    And make it illegal to combine these activities

  9. Steve Randy Waldman writes:

    William — I like very much the quote you excerpt. As in his thinking about uncertainty, Knight was way, way ahead of his time in understanding that wealth is something endlessly reproduced in time, not stuff stored in a warehouse. I had no idea this was his view, but gratifying that he uses the same analogy I frequently do with electrical power (that must be used as generated because it cannot efficiently be stored). I wish economics had learned much more from Frank Knight.

    I also think that the health of bank-like entities is ultimately a function of flows. Bank-like entities are Ponzi schemes. They survive so long as they can generate net inflows of whatever commodity qualifies as “liquidity”, or at least keep flows balanced. If they cannot prevent net outflows over a period of time, they either fail for lack of liquidity, or if they can produce the liquid commodity at will (e.g. a government, central bank, or state-supported banking system), persistent net outflows provoke a devaluation in real terms.

    Accounting capital is a “stock” idea, but its purpose is entirely to be predictive of flows. That is, in theory, a “solvent” bank — one whose assets are clearly worth more than its liabilities, can always generate cash inflows to cover liabilities by pledging or selling assets. As you and Knight suggest, undifferentiated “stock” of undifferentiated wealth is a kind of fiction, but it has been a useful accounting fiction, and is all of what balance sheets struggle to track. But, if you push a fiction too hard, you can exhaust the ability of analogy to stand-in for truth. I think that’s what we’ve done with modern complex banks. The accounting fiction of a “stock” of capital has become too nebulous a concept to accurately capture the ability of a bank to balance flows in extremis, even if it is permitted to slow the pace of redemptions (that is, if it can choose an orderly liquidation rather than sell assets in desperation during a run). The idea of a stock of capital, however artificial, is pretty deeply embedded in conventional accounting. We either need to choose to rescue it by reproducing the conditions under which the fiction mostly work, or come up with something different and better.

  10. Steve Randy Waldman writes:

    Greg — I agree very much in theory that accounting values are at best probabilistic estimates of whatever we intend for them to capture, and as such they’d be better expressed as a distribution (imperfectly via confidence intervals or standard errors, or more richly with a mathematical characterization of the true distribution).

    I don’t think it’s remotely practical for bank accountants do offer this, though. If they did, the same incentives that push bank executives to state capital overoptimistically would push bank accountants to make statistical and econometric choices that generate small apparent standard errors. Given the complexity of existing banks, the “model space” I described is too vast to literally run the thought experiment I proposed. (However, if we forced banks to be very simple, we might “bootstrap” the distribution of capital and other variables by literally sampling the space of regulatation-compliant models.)

  11. Steve Randy Waldman writes:

    glory — I like MM’s piece, and mostly agree that in practice CDS are instruments designed to goad the sovereign into backing them (because the trigger infrequently and all the world treats them as solid, and the state can’t bear to stand aside and let the broad investor class suffer the consequences of risk for which it en masse has neglected to prepare). If CDS become centrally cleared (more a bad idea than good, IMHO), that will complete their transition to public-sector instruments, because a central clearinghouse in the ultimate in TBTF financial institutions.

  12. Steve Randy Waldman writes:

    KD — Yes, the report certainly didn’t proclaim the valuations to be “accurate”. But if a tree misvalues an asset in the forest and no one is there to face a jury, does the distinction between truth and lie really matter? The tolerances that guide executives in valuing their positions are legal tolerances. Regardless of deeper economic truths, valuations that are “good enough” that the law cannot declare them to be false are the valuations whose consequences investors and taxpayers have to tolerate and suffer. “Value” is a social fact, and as social facts are “real” only in terms of social consequences. Misvaluation within a range that provokes no bad consequence for the misvaluer is a social non-fact, a question of Platonic faith in private notions of “true value”. There’s no use getting hung up on what you know to be true when the law does not differentiate that truth from fiction. A corporation is not a person, in some sense, but it can still sue your ass and shelter people who harm you (via limited liability) despite the fact it is entirely fiction.

  13. Steve Randy Waldman writes:

    David — Both the links you’ve cited are in the “Related” section of the post, and one of your excellent AIG articles is linked in the text (re capital stacking and the sort of regulation/valuation games that can be played via crossholdings among related entities).

    I think you and I are broadly in agreement. Note that I don’t in this piece claim that regulatory capital can never serve as a workable basis for regulation, only that modern “large complex financial institutions” are constituted in a manner that makes RBC schemes unworkable. In all of your proposals, you are careful not only to suggest “dumb” capitalization thresholds and charges, but also to demand that banks be made simple enough that those thresholds could be meaningful. You propose restricting what assets regulated financials can hold, preventing structural complexity, and limiting cross-holdings among subsidiaries and affiliated firms. You are quite explicit in suggesting that whatever would muddle the accounting should generally be forbidden of regulated firms. If we adopt that rule and apply it diligently, we’d have standard ways of accounting for the capital of permissible firms and could use that comparable measure as a basis for regulation.

    Dumb capital-based regulation can work, but only for dumb banks. And when I say “dumb banks”, I mean that in a very good way.

  14. […] Capital can’t be measured Steve Waldman. This points to big time problem (IMHO) in what we allow firms to do. If the metrics are that bad, the firms have to be vastly less geared or stick to easier to value products, something no one is prepared to consider. […]

  15. Steve Randy Waldman writes:

    Lasse — That’s about the most concise narrow-banking proposal I’ve ever read. I’ve long been a fan of narrow banking (which is distinct from maturity-matched banking, although Paul Krugman recently misused the term to describe maturity-matched banks). There are big practical issues: the unregulated nongovernment banks may still become too big to fail along a variety of different dimensions, even though insured deposits and the payments system would be isolated from their failure. Bear, Goldman, and the entire money-market fund sector were effectively state-guaranteed, despite no formal insurance. Lehman was the exception that proved the rule, that reminded the state that it must guarantee more of the financial system than it explicitly commits to, or else face severe consequences. To implement narrow banks well, we’d have to 1) insulate the “government banks” — deposits and paymenst — from any exposure to risk investing, AND 2) ensure that the private investing sector is structured so that the failure of any individual investment fund is tolerable, and that in a general downturn, the simultaneous failure of many (but not all) investment funds is tolerable. It’s do-able, but not by accident.

    I still like narrow banking, but I have to admit I’m less certain that its a good idea than I used to be. The reason for my diminished enthusiasm is hinted in my “Can we handle the truth?” post. Banks as presently constituted are effectively a means by which the state invests public money in projects selected by private entrepreneurs and vetted by private bankers. I think state investment in private projects, with state funds directed by private agents incentivized to make commercially sound choices, is an essential component of a modern economy. But we have no political consensus for (and in the US even something of a taboo against) state investment in private, for-profit projects on generous terms. The conjoining of deposits and payments with lending in banks is a way of obscuring the fact that bank-lent funds are in fact state-supplied capital. In a narrow banking world, state subsidies to private capital formation would have to be explicit. I think it’d be a better world, we’d collectively make better choices, if we did move to narrow banking and devised transparent means of subsidizing smart private capital formation. But if we cannot get rid of the fiction that banks invest private deposits until we’ve created a political consensus and alternative institutions to privately direct public capital towards commercially useful projects that unsubsidized private markets would fail to fund. I think without some institution that serves this purpose, “advanced” “capitalist” economies cannot thrive.

  16. Arun writes:

    Forget about regulation for a moment; why would anyone accept Lehman Brothers as a counterparty in any transaction if they cannot evaluate Lehman Brothers’ health? And in general, how does the market work [efficiently] if information is so fuzzy?

  17. Kid Dynamite writes:

    good reply, Steve – now I see your point. I guess the point for me is that we certainly need to be moving toward a MORE accurate asset marking method, not a LESS accurate one, which is what we’re doing in response (FASB 157 modification!). less extend and pretend.

  18. doggyDish Party writes:


    hard” capital and solvency constraints for big banks are better than mealy-mouthed technocratic flexibility. But absent much deeper reforms, totemic leverage restrictions will not meaningfully constrain bank behavior. Bank capital cannot be measured. Think about that until you really get it. “Large complex financial institutions” report leverage ratios and “tier one” capital and all kinds of aromatic stuff. But those numbers are meaningless. For any large complex financial institution levered at the House-proposed limit of 15×, a reasonable confidence interval surrounding its estimate of bank capital would be greater than 100% of the reported value. In English, we cannot distinguish “well capitalized” from insolvent banks, even in good times, and regardless of their formal statements.

    Lehman is a case-in-point. On September 10, 2008, Lehman reported 11% “tier on

    Seen from a great distance, our industrial revolution covers hundreds of years. From wide historical perspective easy to visualize the changing significance of investment. Originally investment was the money of feudal war lords with excess sacks of gold betting on the new assembly line machinery about to become financially self-sustaining.

    Is our war-lord’s-gold tank now running on empty? Is the plethora of new widget designs an oversupply on the event horizon or falling through a spiral to the bottom? “Do I hear zero? Sold to the man who bid ‘approaches zero’!” Are tier one, hard assets changing to scrap-metal-value as the news of obsolescence hits the wire? Has the multiplier-effect morphed into the fragment-effect shattering first the most highly leveraged who failed to sell their Ponzi shares in an orderly, front-runner, timely manner? Has our gigantic beautiful Adam-Smith Fern fallen into the peat-bogs where it is crushed by its own ponderous complexity, its own ponderous weight?

    U B Seer

    U B Judge

    U B Thurgood

  19. […] “Regulation by formal capital has a proud and reasonably successful history, but has been rendered obsolete by the complexity of modern financial institutions.”  (Interfluidity) […]

  20. minderbender writes:

    “On September 25, 2008, Lehman declared bankruptcy.”

    September 15, 2008 is when Lehman Brothers Holdings Inc. commenced its chapter 11 case.

  21. As in #8, above: Why don’t we turn the banks into regulated public utilities? Then if the rich want to go off and play the ponies, let ’em. Just not on our dime.

  22. Steve Randy Waldman writes:

    Arun — Large complex financials have innumerable counterparties, almost none of whom even attempt any sort of independent evaluation of firm solvency. Generalized non-specific “confidence”, including the unthinkability of failure, is a core asset. Capitalism has never figured out any other way of doing business. With or without formal state support, bank-like organizations have always relied upon creating an aura of invincibility, and customers have relied upon that aura — and on the false assurance of demandable deposits and other guarantees that cannot be met in aggregate — as a proxy for rational evaluations of solvency.

    Goldman has lots of counterparties. It also has tone of “Level 3” assets whose value no one can independently verify.

    An account of business that treats financials as just like other business organizations whose suppliers, customers, and counterparties vet for creditworthiness in the course of their collaborations is elegant, but also plainly false. People trusted Lehman for the same reasons, and with as good information, as they trust Goldman, JPM, Citi, BAC, or Wells today.

  23. Steve Randy Waldman writes:

    KD — For the moment, we are moving towards a world in which bank financials become ever more pro forma, and bank solvency is ultimately entirely a matter of indefinite liquidity support by the state, combined with obscure subsidies and ex post tilting of investment outcomes if necessary to restore confidence to indispensable organizations.

    Graveyards are filled with indispensable men, they say, but we seem less and less inclined to put our indispensable banks there, “resolution authority” notwithstanding.

  24. Steve Randy Waldman writes:

    doggyDishParty — Your comments remind me of Shirky’s recent essay on collapse and complexity.

    Thomas Jefferson thought we’d need a revolution every 20 years. One way or another, systems grow corrupt, complex, and dysfunctional, and then must “collapse” and be replaced. But nothing says the revolutions must be bloody or the collapses indiscriminately destructive. The real argument about financial reform is between those of us who would like to see a controlled demolition and replacement of what we see as an impossibly dysfunctional system, and those who want to patch it a bit and carry on, who think it can continue (and who perhaps benefit from its continuance) without collapse.

  25. Steve Randy Waldman writes:

    minderbender — Thanks! Corrected, with attribution at the end of the post. I don’t know how I got that wrong!

  26. Steve Randy Waldman writes:

    lambert — I’d love to see us give a try to public no-risk narrow banking combined with a private never-too-big-to-fail financial sector. But per my comments in 15 above, I think we’d find that the level of investment and risk-taking that would occur without public subsidies hidden in the banking system would be too low. One way or another, we need privately directed public investment, but we need much better and sharper incentives for the directors of that public capital (bankers, in our current system). Ideally, we’d make the role of public capital in “private” capital formation explicit, we’d transparently and democratic shape it. But we are all at ideological loggerheads, none of which are consistent with the actual practices that give life to our economy. If we described it accurately, from the left and from the right and every other direction, we’d find the golden goose unacceptable and strangle it. But, by obscuring the real nature of our economic practices, we leave them weak and corruptible and destroy them in a different way. Which is what we have done.

    If we don’t understand how our system (sometimes) “works” — and we are unwilling to understand it — we do a crappy job of keeping it in working order.

  27. […] on | April 4, 2010 | No CommentsTaking as an example the 2008 bankruptcy of Lehman Brothers, Steve Randy Waldman at Interfluidity examines the difficulty of measuring the solvency of major fina… that deal in complex instruments.It’s that “Knowledge Problem” that Professor […]

  28. […] Capital can’t be measured – via Interfluidity – Simon Johnson and James Kwak are absolutely right. Sure, “hard” capital and solvency constraints for big banks are better than mealy-mouthed technocratic flexibility. But absent much deeper reforms, totemic leverage restrictions will not meaningfully constrain bank behavior. Bank capital cannot be measured. Think about that until you really get it. “Large complex financial institutions” report leverage ratios and “tier one” capital and all kinds of aromatic stuff. But those numbers are meaningless. For any large complex financial institution levered at the House-proposed limit of 15×, a reasonable confidence interval surrounding its estimate of bank capital would be greater than 100% of the reported value. In English, we cannot distinguish “well capitalized” from insolvent banks, even in good times, and regardless of their formal statements. […]

  29. Thanks Steve,

    Well stated, particularly the part about shady errors in reported capital being so porous that the “confidence interval surrounding those estimates is greater than 100%.”

    That is a stunning way to summarize the abuses heaped upon the public and long term shareholders

  30. carping demon writes:

    “regulatory postmodernism”

    I’m so glad to hear someone else say that. It nails the academic side of this.

  31. […] Can capital be measured? – Ed: warning severe econogeek alert –  (Interfluidity) […]

  32. dogDish Party writes:


    doggyDishParty — Your comments remind me of Shirky’s recent essay on collapse and complexity.

    Thomas Jefferson thought we’d need a revolution every 20 years

    ~~Steve Randy Waldman~

    Vraiment

    Accomplished writers imitate. Unique writers plagiarize. Was I robbing Shirky or Shirky Jefferson? One thing for sure — Jefferson Memorial was sure popular this Cherry Blossom Weekend.

    Ciao

    dD P

  33. The problem with “no risk” (that’s a misnomer, of course) narrow banking is that you are effectively draining capital out of the economy. Even without the ludicrous constraint of “100% reserves” (i.e., a government-run safe deposit box that apparently would be conducted at an enormous loss, so that it could pay at least minimal interest to depositors even though their money would be idle), any money deposited into the narrow-banking system would effectively be hidden under the mattress: it would not be available to the rest of the economy, could not finance any expansion, would probably guarantee zero to negative growth in the overall economy, and would never accumulate fast enough to permit savers to retire, for example. This is exactly the liquidity trap Keynes was trying to combat with the remedies proposed in his General Theory. Savers were so frightened by the mid-‘thirties that they were refusing to make any capital available to allow the economy to return to productive levels. Why should we ever want to go back there??? Surely there are better cures for a misrun banking system than one that kills the patient!

  34. thabks This is exactly the liquidity trap Keynes was trying to combat with the remedies proposed in his General Theory. Savers were so frightened by the mid-’thirties that they were refusing to make any capital available to allow the economy to return to productive levels. Why should we ever want to go back there??? Surely there are better cures for a misrun banking system than one that kills the patient!

  35. Indy writes:

    Superb posting, as usual. There’s a military-science “composite risk-management” analogue about needing to expend significant real resources to control generalized entropy (narrow the bell curves in all relevant dimensions, build walls, that kind of thing), an effort which yields the tiniest ability to meaningfully analyze the long-term strategic impact of short-term alternative courses of action in a chaotic universe with seemingly infinite states of nature. But I won’t get into all that.

    Reminds me a bit of the CBO-scoring process / Social Security Trustees Solvency report / general long-term government estimates category of problem – which, come to thing of it, isn’t so much different from a long-term exotic capital valuation modeling problem. Also economic cost-benefit analysis of long-term climatology / global warming. The error bands dwarf the scale of the mean result. You end up with substantial possibilities of significant deficit or surplus, or of an increase in agricultural output or mass extinction.

    These things simply cannot be made conveniently predictable. Precaution against catastrophe would have to include some kind of compartmentalization, firewalls, and forced diversity – to prevent all the small individual instabilities in the sand-dune from getting connected with continuous fingers where tiny shocks quickly collapse the whole thing in a sudden avalanche.

  36. flow5 writes:

    The commercial banks used to “store” their liquidity & not “buy” their liquidity. This began about the time Citibank invented the negotiable CD. The politicians let the bankers buy what they already owned (pay interest on their customer’s deposits). Thus the banks, & the banking system, are less profitable, long-term interest rates are much higher, & the economy is much slower, etc.

    The banks should be severely circumscribed in both their assets & liabilities (i.e., severely limit financial innovation or the loan & investment manager’s trial & error). Historically the only way this works is in a “command economy”. So the only way to discipline the banks will be to nationalize them. In this case, the slower government bureaucracy, the better.

  37. nadezhda writes:

    No quarrel at all with your analysis (and that of the commenters) re the pitfalls and uselessness of much of FI accounting. It’s not just the regulators and the counterparties who can’t get a meaningful grip on the “true” financial condition of an FI from their balance sheets. Even if an FI isn’t gaming the system, these structures have become too complex for managers to understand the risks in the businesses they’re running.

    So we impose some “dumb” rules that help produce simplified “dumb” banks out of the baroque messes we currently have. But that puts us back where all of this started three or four decades ago. Disintermediation from the depository institutions into the money and capital markets. The banks (and S&Ls) couldn’t compete, so they pushed their way into lots of riskier stuff (recycling petrodollars, one real estate bubble after another, sovereign debt and so forth) and worked loopholes, cross-border holdings, off-shore subsidiaries, and off-balance-sheet vehicles to play in areas where disintermediation had moved much of the business. By the time Glass-Stegall was abolished, it was already pretty much dead as a practical matter. What makes us think it would be any different if we reimposed a Glass-Steagall-type system — whether via “dumb” rules that in effect would constrain banks from working across too many speculative business lines or via attempts to create a “narrow banking” regime?

    I simply despair at putting the genie back in the bottle. Even if we could prune back the monster depository institutions we’ve allowed to develop, we’re still going to have financial bubbles and panics. But if the banks are pruned, even more financial activity will be in a mostly-unregulated global financial system which would be nearly unregulatable without a global financial regulator. At least now with the banks being significant players in the global markets, each home country central bank for each of the regulated behemoths takes some responsibility for lender-of-last-resort and resolution functions when one part of the global markets or another blows up.

    So though I don’t oppose getting tougher on regulating the banks — both in terms of the rules they have to comply with and the way we measure whether they are complying — I think this has to be accompanied by trying to bring the money and capital markets into a regulated environment. Whether it’s via rules for membership and instruments traded on exchanges and via clearinghouses, or imposing fees (akin to depository insurance) on non-bank FI liabilities to level the funding playing field, or even outlawing some types of financial instruments, somehow the markets and shadow banking system have to come within some form of regulatory discipline and accountability.

    For example, I take Steve’s point that pushing derivatives into a clearinghouse creates another TBTF situation. But in the absence of a clearinghouse, and even if we force the banks out of the derivatives business, the government is going to have to bail out the derivatives market in any event if it’s a key part of a bubble or becomes a major transmission belt for contagion in the wake of a crash in another part of the financial markets. So IMO, better that we bring those activities within a more transparent system subject to enforceable rules, even though we run the risk of having to bail out the clearinghouse. And btw, because of financial incentives, clearinghouses are industry-sponsored “self-regulatory” institutions which have a rather successful track record of rulemaking and disciplining members who try to run their business at the expense of the clearinghouse or other members. Where “self-regulation” tends to be extremely lousy is in rulemaking and disciplining members re their behavior with clients, where the financial incentives for regulating members’ behavior are feeble at best.

    In any event, IMO there’s been an excessive attention to TBTF as if that’s the key to reducing the incidence of crises or eliminating the need for bailouts. As Krugman pointed out (this weeked?), TBTF is important. But financial panics that require the intervention of a lender-of-last-resort function happen all the time even in the absence of TBTF institutions. The bigger problem with TBTF is that it can make rescue more costly, reduce the effectiveness of the cleanup process, and make it politically difficult to stick the guys who created or facilitated the mess with (part of) the cost of the bailout. So I’m all in favor of putting a cap on the size of individual banks and doing what we can to counter the excessive concentration in the banking industry. But that’s more a political economy issue of an unaccountable plutocracy, and less a matter of regulating risk or reducing moral hazard.

    I fear that if we try to stick the banks back into their box without also taking on the money and capital markets, we’re just going to see more and more activity and “financial innovation” gravitate to the unregulated markets. But we won’t have reduced the necessity for governments to bail out the system when the inevitable boom-crash happens again — in fact we will have added even more unaccountable risk to the system, increasing the odds that the next bailout will be even bigger and messier.

  38. Sam writes:

    Former economics grad working on Wall St – if there is one place we should be able to measure capital it is a financial firm. No heterogeneous inputs etc. aka Cambridge controversy. Money is the measure of value but here money is the only value.

    The problem is that the people currently working on Wall St are completely incapable of accounting for their assets and liabilities in a swaps world. In one sentence – the accountants on Wall St know no statistics. None.

  39. […] The difficulty of measuring leverage of banks"…for large complex financials, capital cannot be measured precisely enough to distinguish conservatively solvent from insolvent banks, and capital positions are always optimistically padded." […]

  40. Steve Randy Waldman writes:

    Andrew & sexsexsexsexsexsex.com — The trouble with your critique of narrow banking is that depositors don’t actually supply capital any more. They don’t bear risk (which is the heart of what it means to supply capital), and they allocate liquidity for reasons that have nothing to do with the quality of investment opportunities available to banks.

    In reality, nearly all bank capital is now supplied by the state. The portion of their balance sheets funded by depositors is formally insured (for the most part), and uninsured creditors of large commercial banks clearly have a substantial fraction of their investment risk tacitly absorbed by the state. Modern banks are narrow-banks-within-banks. Depositors supply funds that might as well be put under mattresses, and then the government supplies risk capital. The competition for deposits affects banks’ costs of funds (retail deposits are cheap), but not their access to capital (in the sense of funds they are able to invest for profit and at risk). For large banks, the only meaningful constraint on access to “loanable funds” is their regulatory capital position, which, if you buy my headline rant, is itself gamed and gamable. Banks try to circumvent regulatory capital requirements because they are the only obstacle that prevents them from putting public funds at risk for private profit.

    If that sounds like a critique, it’s not. I think that we absolutely need institutions whereby commercially motivated actors direct public capital towards economically valuable uses, and are rewarded if they direct the commonweal well, and punished if they take bad risks. For such institutions to work, the state must diversify its reliance upon capital allocators and limit the capital available to each (so that one corrupt or foolish organization can’t exact terrible costs). If regulatory cap constraints are circumventable, the state finds it difficult to do so.

    Anyway, with respect to narrow banks, you are right that of we just turned existing banks into glorified mattresses without creating alternative institutions by which private agents make commercial investments on behalf of the state, we’d have a terrible underinvestment problem and a miserable economy. Any narrow banking proposal has to be combined with investment funds proposals that would serve the role banks presently serve, and my view is that those investment funds would have to be subsidized. That sounds radical, but it is just rendering more transparent our existing system. And with transparency, we might better ensure accountability and exercise control.

  41. Steve Randy Waldman writes:

    Indy — I think you’re thinking about the banking system in exactly the right way — as a complex system given to unpredictable fragility of not carefully arranged. Like any complex distributed system, “arrangement” has to simultaneously leave room for creative and autonomous operation of parts, but be designed in such away that problems, failures, and catastrophes that occasion its pieces do not spread and take down the whole.

    If one applied the sort of military standards you are describing to the banking system, it would be found laughably wanting. Although I don’t think we should organize the banking system along the lines of a military operation per se, paranoid attention to robustness and redundancy would be a wonderful thing to incorporate.

  42. Steve Randy Waldman writes:

    flow5 — Banks in a sense already are nationalized, since the capital they invest is almost entirely public capital. They are pretend-private, though, on the theory that private-sector-style, commercially motivated allocators of state capital would do a better job than explicitly political bureaucrats. But banks are regulated too, a recognition of the fact that the consequences of their risk-taking are not ultimately private. Where on the spectrum between public bureaucracy and fully private firm banks ought to sit is a hard question. I’m still free-market-ish enough that I don’t want permanent nationalization, the-slower-the-better. But I do want more explicit acknowledge that banks allocate public capital, and much more intrusive management by the state of how its arrogant managers put together our aggregate capital portfolio. Getting right the mix between accountability and broad objective set by the state and commercial “private” allocation pursued by profit-seeking bankers is what we are struggling to do.

  43. Steve Randy Waldman writes:

    nadezhda — I don’t have much to dispute in what you say.

    I would clarify that my misgivings about clearinghouses are mostly confined to CDS, whose jump-to-default property makes them particularly hazardous. Futures and options markets have worked remarkably well, and as long as we ensure the managers of clearinghouses are incentivized to set conservative margin requirements (ideally by having them organized as partnerships rather than limited liability entities), they do a lot of good. I agree with David Merkel that we should move interest rate swap to cleared exchanges and get some experience with those, and them consider more exotic products. (I ultimately don’t think CDS deserve to live, because the margins would have to be set so conservatively that people might as well buy or short the bond, and because they dangerously separate capital provision from control rights. But if someone can show me a safe way to margin them that keeps them sufficiently liquid, I might revise my opinion.)

    Ultimately, as has become fashionable to say, I think the dichotomy between TBTF and other sorts of regulation is overstated. Part of why big is bad is because it makes prudential regulation impractical (because big is too powerful to regulate). Smallness and supervision are complements, not substitutes. I don’t think that “large complex financial institutions” on the scale that we have can be successfully regulated for a long period of time. With respect to the bad alternatives we might create if we squeezed incumbents, shadow banks in capital markets, etc., I don’t think the goal should be to try to make every imagined financial arrangement regulated and safe, but to ensure that whatever sort of bank-like or market-like entities are tolerated are EITHER outlandishly safe OR can fail without widespread adverse consequences. Shadow banks can securitize all they want, as long as the holders of the securities are purely private individuals willing and able to bear the risk they sign onto. Regulatory circumventions via capital markets become hazardous when institutions that cannot fail gracefully invest in them, or bear residual risk by virtue of originating them.

  44. Steve Randy Waldman writes:

    Sam — Great point, one I hope to expand on someday. As you say, financial firms should in fact be unusually simple. Their assets are homogeneous in many respects, and denominated in the conventional unit of account. Reorganizations of financial institutions are nothing more or less than reshufflings of ownership of assets, very few of which are bound to particular people or places. (The owner of a factory may be irreplaceable, in the sense that no one understand how the machinery works without her. But the owners of financial assets are as fungible as their holdings. There may be exceptions for complex trading positions, but there’s little evidence that any social good comes from those exceptions. Also, when “relationship banks” hold and monitor illiquid loans, there may be a synergy between a particular staff and the assets it holds. But that has become less and less true as even commercial banks have become large and replaced “soft information”-based lending with “hard” proxies like fico scores and simple collateral valuation.)

    Anyway, I won’t comment on what people on Wall Street do or don’t know — you’re in a better position to judge. But I am pretty sure that degree of complexity modern banks create and then fail to manage is not necessary, or at least needn’t be organized into agglomerations are beyond human analysis but the consequences of misjudgments are to terrible to bear.

  45. nadezhda writes:

    Misc reactions to your response.

    1. I too would love to kill CDS, but bringing them out of the OTC environment as well as within a clearinghouse would go a long ways. Lots of the CDS world would turn out to be uneconomic if standardized, subject to transparent, continuous re-pricing, and forced into a clearinghouse regime that imposes the full price of contingent exposures. But that means severely restricting the amount of bespoke business that can be done, which will be tenaciously opposed by Wall Street.

    2. Re ownership structure of a clearinghouse — the problem isn’t limited liability, it’s “mission”. A clearinghouse shouldn’t be a profit-maximizer for return on shareholder investments. It has to be an industry utility, where owners get most of the economic benefits of their investment through obtaining safe, efficient, cost-minimizing services. The presence of outside shareholders introduces competing business purposes, encourages short-termism and window-dressing of quarterly financial reporting, and produces conflicting and misaligned incentives. That, far more than limited vs unlimited liability, is why “going public” by the investment banks has been so pernicious. (And I’d also limit outside ownership of exchanges and other trading markets. The objective of managing an industry utility for safety and efficiency will inevitably conflict with maximizing ROE and producing quarterly results to please the trading markets in which managers realize the bulk of their compensation via options etc.)

    3. I’m not concerned about securitization as an area requiring special regulation. We actually have a regime — securities law — that is available to deal with securitization. The regulatory process failed us this time around, but now that the potential hazards of that type of instrument are widely appreciated, it will be awhile before system-threatening risks build up in that area again. For the regulated banks as distinct from the non-bank FIs, the core problem wasn’t their participation in the securitization process per se, it was failure to recognize the residual risks embedded in the dodgy assets they were busy manufacturing. If we actually enforce the rules governing securitization — both the instruments themselves via securities law and the banks as manufacturers of those instruments — we don’t need any special rules to extend into the shadow banking system. We just need to strengthen a few rules (roles of rating agencies, off-balance-sheet vehicles) and enforce the existing rules. (Yes, I know, famous last words and all that, but at some point we have to assume at least some regulatory capacity at the margins that disciplines the dodgiest stuff).

    4. The objective of bringing more of the money and capital markets into a regulatory regime isn’t to make all financial assets “safe”. I’m concerned about trying to limit what the big banks can do without some sort of limits on the liabilities side of the non-bank FIs. If we try to put the genie back in the bottle, IMO we’re going to increase the scope and probability of more LTCMs, no one of which on its face looks TBTF, but which can produce disastrous contagion.

    Decades ago, before the process of disintermediation got going in earnest, the banks could be conservative because, as a group, they had an effective monopoly on deposit-type liabilities. But when they had to compete with the markets for funding, they felt compelled to move up the risk ladder on the asset side as well. Unfortunately, their track record shows they’ve been pretty lousy at it. That’s certainly an argument in favor of trying to force the banks to stick to their old knitting and leave the markets to innovate and manage capital allocation and risk. But if we’re going to limit what the banks can do on the asset side, and if we’re going to protect against TBTF banks by imposing additional funding costs on the big banks (in effect to extend the cost of FDIC-type rescue and clean-up beyond the deposit insurance system and place a penalty on growth for its own sake), we’re further disadvantaging the banks competitively vis a vis the markets. The result would not be a partial reversal of disintermediation but a further acceleration of that process.

    Now, I take it your response would be “Great, the banks should be utilities, not risk-taking institutions. It’s the markets, where the rule is caveat emptor, and everyone knows that there’s not a government bail-out backstop, where we want capital allocated and risk lodged.” But that, it seems to me, is naive.

    When the next bout of irrational exhuberance appears (and one inevitably will), a prominent feature of the bubble will equally inevitably be the manufacturing of lots of dodgy financial assets and erection of fragile towers of credit to exploit the bubble. And if most of our allocation of risk capital has been shifted away from the banks and into the capital markets, we’ll have few mechanisms to put on the brakes other than the blunt instrument of monetary policy to try to sop up excess credit creation and liquidity.

    So we’ll eventually have another system-wide run on the money markets when the bubble bursts, and another crash of asset prices and wealth destruction, with its inevitable impact on the real economy. And if you think the central banks had a hard time figuring out how to play lender-of-last-resort for the money markets during the 2008 panic, it’s going to be even harder if they have to play lender-of-last-resort when the contagion is spreading through lots of smallish shadow banking institutions with which they have few regulatory or oversight relationships.

    Central banks and governments will have even fewer mechanisms to manage the rescue and clean-up process because the mess will be scattered across a range of institutional structures, legal and regulatory jurisdictions, etc. Because the markets are so fluid and truly operate globally, fixing which central bank is responsible for what institutions and markets, and coordinating multiple lender-of-last-resort actions to defeat contagion would be even more of a nightmarish collective action problem than it was this time around.

    Now, since I view a global financial regulator as neither feasible nor desirable, if we’re going to treat banks as utilities and rely instead primarily on the markets for capital allocation, we need to give some thought to what sort of extensions of national regulations are necessary to reduce the frothiness of the markets and to improve the ability of national central banks to perform the rescue and cleanup functions that will inevitably be required in the next bubble-and-bust episode.

    5. I take your point about the overlaps between TBTF and safety-and-soundness concerns. Yes, putting a ceiling on the size of any one bank would also put some sort of cap on the institutional complexity which has outstripped the capacity of managers to understand and manage. So size matters, both at the institutional level in terms of capacity to manage risk, and at the system level in terms of political economy.

    But we also have to acknowledge what getting rid of TBTF institutions won’t achieve — and here’s where I think the obsession with TBTF has done a disservice to the conversation about “what should be done.” It wouldn’t in any way eliminate the need for the public to bail out the financial system when the inevitable bubbles and crashes occur. What we should hope to achieve by limiting or eliminating TBTF institutions is a more orderly rescue-and-cleanup process that has less impact on the real economy. And my point re the capital markets is that by trying to get rid of TBTF banks and shifting capital allocation and risk to the markets, we are likely to increase rather than reduce the cost of rescue-and-cleanup, both in terms of the price of the bailout and the impacts of the financial crisis on the real economy. So we’re going to have to start what I believe has been carefully ignored to date in “financial reform” discussions — think about how to extend rules, regulatory oversight and central bank cleanup functions to the global capital markets.

  46. […] by UnrealI was very struck by a piece by Steve Randy Waldmann at Interfluidity yesterday, entitled Capital Can’t be Measured.  He is basically arguing that modern financial institutions are sufficiently complex that the […]

  47. Steve Randy Waldman writes:

    nadezhda — Although we’ve points of commonality, we’re likely to disagree on the core TBTF issue. We agree that we must assume that we will never eliminate financial crises. But I think that large, complex institutions make them more likely (due to incentives of employees to trade huge short-term payouts for long-term organization risk, game theory incentives of manager and creditors given the social cost of falure, the efficiency with which errors achieve large scale, complexity that makes errors difficult to perceive and avoid, and the ability of political power to neutralize regulation), that LCFIs increase the intensity of financial crises, make them more difficult to manage in the immediate term (i think you badly overstate the problems with mitigating the costs of a financial crisis dispersed among small entities), and that LFCIs make crisies nearly impossible to manage over the longer term, in the most important sense of maintaining good incentives for capital allocation in between crises. (Remember, the financial crisis is the frosting that sits on top of a cake of tragic misuses of capital over a long period of time. I care far less about the financial crises than I do about the anesthetized costs of the bubbles that precede them.)

    I think you think that, since TBTF institutions can be identified and watched, we can use regulation to diminish the likelihood and severity of financial crises (and to manage the perverse, gargantuan incentives of agents whose expected lifetime net worths are dwarfed by potential bonuses on a particular deal?), that TBTF-centered crises are not more intense than other crises, and that they are easier to mitigate. (I don’t know if you have a view on their effect on the long-term quality of capital allocation.) If I’m not mischaracterizing (I might be — this is rushed!) I disagree with you on all counts.

    But that’s what keeps this fun!

    BTW, I think the main costs of crises and bailouts are not their financial costs, but (repeating myself) their effect on incentives going forward and (this is new) the unequalness and legitimacy of the distribution of wealth looking backward and in real time. Again, small helps mitigate those costs, as fewer people affiliated with smaller orgs get dynastically rich based on one-shot, perhaps-economically-stupid cash-outs.

    I think our views on derivatives and clearinghouses are more similar than different. I didn’t mean to make a fetish out of limited liability — it was just an example intended to make sure that operators of clearinghouses don’t every try to treat them as options whose catastrophic downsides could be sloughed off to others. But there may be lots of ways to effectively run clearinghouses and “public utilities”, and as long as they are viewed as risk not profit centers and we avoid the sort of dispersion of consequences you worry about, whatever works is fine with me. (I do think unlimited liability helps!). Re CDS, we also mostly agree. If clearing them is workable without putting clearinghouses at risk, if clearinghouses are string enough to resist strong calls to let them be undermargined given their gap-to-default hazards, then they’d be better cleared than not. Those “ifs” are pretty big though — I really don’t want to see a clearinghouse failure, so before we clear CDS, I want to be sure that we have institutions strong enough to resist calls for CDS on status quo ante leverage terms, which would be fatal.

  48. Steve Randy Waldman writes:

    nadezhda — BTW2, I don’t think banks should be utilities (at least not banks that lend). I think they should take risks, but they should be regulated to prevent herds of banks from taking correlated risks, and they should be small and well-capitalized enough so that the costs of uncorrelated failures are mostly private. I like banks-as-relationship-lenders, taking risks based on client specific “soft” information. I think that’s a very clever sort of institution, although whether they should be deposit-funded is a separate concern.

  49. Steve Randy Waldman writes:

    nadezhda — Sorry if I’ve mischaracterized your thoughts or failed to address your points specifically. I do have to run, just read your text and gave a quick reaction, and now I’m looking back and feeling like I’ve not treated your specific views with due care! Sorry (but gotta go)!

  50. […] was very struck by a piece by Steve Randy Waldmann at Interfluidity yesterday, entitled Capital Can’t be Measured.  He is basically arguing that modern financial institutions are sufficiently complex that the […]

  51. […] (8% Tier 1 capital, but only half of that has to be tangible equity). But still, is it enough? As Steve Randy Waldman pointed out, Lehman turned out to be worth between $50 and $160 billion less than its books said it […]

  52. […] (8% Tier 1 capital, but only half of that has to be tangible equity). But still, is it enough? As Steve Randy Waldman pointed out, Lehman turned out to be worth between $50 and $160 billion less than its books said it […]

  53. The idea that people’s only choice is to deposit their money in a bank that will use it to fund private corporations’ growth is somehow like socialism for the rich. Same way as people’s pension savings are basically tied in different commercial papers and public stock. Common wealth is by law and without choice used as tool for banks making money, while the original owners get only a scrap of the profits, and ultimately still holds the risk if these bank activities go south. In other words, all nation’s people involuntarily participate in means of production and resources with use of their capital.

    In this way I am a bit sceptical about Steve’s point about public funding of private projects. At least in Finland they are usually very inefficient, bureaucratic, and many times are a perfect example of Friedman’s description of do-gooder/special interest unholy union, where the benefit goes to a third party.

    I am at least in favor of a choice that people could decide if they want to deposit their money in government 100% reserve banks, or private banks that use that money in their various financial activities.

  54. […] Capital can’t be measured – interfluidity […]

  55. […] Randy Waldman so I don’t want to cricitise him too much, but I think he makes an error in the following: On September 10, 2008, Lehman reported 11% “tier one” capital and very conservative “net […]

  56. […] from us, and blow up the bank.” But think about this: this type of error only goes one way. Steve Randy Waldman has already made this point about capital:  “For large complex financials, capital cannot be […]

  57. […] Steve Randy Waldman explains why measuring bank balance sheets (capital) is impossible […]

  58. Claud Rast writes:

    I can’t resist leaving comments.