The 2008 crisis: markets, regulation or politics?

Suppose you are the finance director of a group presenting a risk review to the board, and a director asks whose fault 2008 was. Three sources give three answers. This post sets them side by side, marks where they agree and where they collide, and names no winner, because the sources I read do not support one. The way in is the library's case on Lehman Brothers and the wider US financial system, 2007–09.

Where does each source put the blame?

The FCIC stresses supervision and corporate governance, Acharya and colleagues stress state guarantees and the way rules measured risk, and Tarr stresses political decisions about housing.

The FCIC had ten commissioners: six adopted the report and four dissented in two separate statements (Financial Crisis Inquiry Commission, 2011).

How much of the crisis belongs to Fannie Mae and Freddie Mac?

The sources give no common figure, and the reason is how each one measures. Tarr puts the failure to regulate the two government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac, at the top of his list of regulatory failures (Tarr, 2010). The FCIC concludes they "contributed to the crisis, but were not a primary cause" and "followed rather than led Wall Street" into risky lending (Financial Crisis Inquiry Commission, 2011, p. xxvi).

Much of the disagreement sits in the underlying data. Tarr relies on Pinto, a former chief credit officer of Fannie Mae (Tarr, 2010), and Acharya and colleagues cite Pinto's testimony for their 15 per cent (Acharya et al., 2011, p. 15). Wallison, who dissented from the FCIC, says his estimates of subprime and Alt-A loans come, unless he notes otherwise, from Pinto, and that key mortgage-market definitions have never been fully agreed: a FICO score below 660, or a label given by the originator (Financial Crisis Inquiry Commission, 2011, p. 451). I did not check Pinto's data myself.

On the Community Reinvestment Act (CRA, 1977) the clash is direct. Tarr argues that its new enforcement in the mid-1990s pushed banks to lower lending standards (Tarr, 2010). The FCIC "concludes the CRA was not a significant factor in subprime lending or the crisis", noting that only 6 per cent of high-cost loans had any connection to the law (Financial Crisis Inquiry Commission, 2011, p. xxvii), and Hennessey, Holtz-Eakin and Thomas agree (Financial Crisis Inquiry Commission, 2011, p. 414).

On mechanism, the two sides are closer. The FCIC calls the GSEs' business model "deeply flawed": listed companies with implicit government backing (Financial Crisis Inquiry Commission, 2011, p. xxvi). Acharya and colleagues use the same two firms as their example of unpriced implicit guarantees (Acharya et al., 2011, p. 14).

Did deregulation cause the crisis?

The three answer differently, partly because they attach the label "deregulation" to different things. One decision by the US Securities and Exchange Commission (SEC) in April 2004 shows it.

Under a voluntary supervision programme, the SEC let the five largest investment banks use their own value-at-risk (VaR) models to calculate required capital. It estimated that average capital charges on securities portfolios would fall by 40 per cent, and Lehman Brothers, commenting on the proposal, wrote that it "applauds and supports the Commission" (Financial Crisis Inquiry Commission, 2011, p. 152).

Three sources, three names for it. Acharya and colleagues, who date the amendment to August 2004: the investment banks lobbied the SEC, and the net effect was essentially to double the gearing of investment banks (Acharya et al., 2011, p. 13). Tarr: not deregulation but the imposition of internationally coordinated Basel standards, though he counts it among the regulatory failures (Tarr, 2010). The FCIC tells both halves: the banks did lobby (Financial Crisis Inquiry Commission, 2011, p. 151), but leverage ratios had been higher in the late 1990s, so the programme was not solely responsible (Financial Crisis Inquiry Commission, 2011, p. 153), even though it was widely viewed as a failure (Financial Crisis Inquiry Commission, 2011, p. 154).

In places the sources talk past each other. Tarr lists three decades of deregulation as removing interest-rate ceilings, easing branching limits and letting commercial banks into underwriting and insurance (Tarr, 2010). The working paper I read does not mention the 2000 law that barred regulation of over-the-counter derivatives, which the FCIC calls a key turning point (Financial Crisis Inquiry Commission, 2011, p. xxiv). Wallison also rejects the deregulation story (Financial Crisis Inquiry Commission, 2011, p. 445), while Hennessey, Holtz-Eakin and Thomas reject both poles: the amount of regulation is the wrong measure (Financial Crisis Inquiry Commission, 2011, p. 414). Acharya (2013) changes the question to the kind of regulation: Basel capital rules were in place, but they looked at each institution's health, not at aggregate risk (Acharya, 2013).

Where do the three sources agree?

On mechanisms, even where they differ on weight. The right-hand column is the question I draw for your own business; the sources do not pose it.

Coverage is uneven: Tarr mentions short-term borrowing only in passing, and Acharya and colleagues raise ratings in one recommendation. The table suits a list of questions, not settling policy.

Was Lehman Brothers too big to fail?

Here the sources collide most directly over an event. Tarr writes that Lehman's bankruptcy showed fear of systemic collapse to be exaggerated: its very large credit default swap (CDS) holdings were unwound within four weeks, with every party paid on the original terms (Tarr, 2010). Acharya and colleagues tell a different story: the fall in the value of Lehman's short-term debt made the Prime Reserve Fund break the buck, setting off a run on the whole system that only a 100 per cent government backstop of money market funds stopped (Acharya et al., 2011, p. 17).

Tarr examines CDS settlement; Acharya and colleagues, money funds and short-term funding markets, which Tarr's working paper does not discuss. The FCIC looks at how the government behaved: it rescued Bear Stearns, let Lehman fail, then saved AIG, and that inconsistency added to uncertainty and panic (Financial Crisis Inquiry Commission, 2011, p. xxi).

How should you read these accounts without picking a side too early?

Ask two things before trusting a number or a conclusion.

Denominator. Take AIG alone: the FCIC records $79 billion of protection sold to investors in mortgage-related securities (Financial Crisis Inquiry Commission, 2011, p. xxiv), Acharya and colleagues $450 billion of one-sided CDS exposure on AAA tranches (Acharya et al., 2011, p. 16), and the 2013 article over $500 billion in notional insurance (Acharya, 2013). Three figures, three bases I cannot reconcile from these texts. The post on relative and absolute risk makes a related point about bases.

Cure. Each cause implies a different medicine. The FCIC was not charged with making recommendations (Financial Crisis Inquiry Commission, 2011, p. xxvii) but calls the Federal Reserve's failure to set prudent mortgage-lending standards the prime example (Financial Crisis Inquiry Commission, 2011, p. xvii); Tarr would subsidise down payments as Australia does instead of forcing banks to lower standards (Tarr, 2010); Acharya and colleagues would price the guarantees and charge for systemic risk (Acharya et al., 2011, p. 2).

Going deeper: what are these three sources, and how much did I read?

The FCIC. I read the Conclusions (pages xv–xxviii), pages 151–154 on the SEC programme, and the two dissenting statements (pages 413–419, 430 and 445–451). I did not read the rest of the majority report, including the chapter on Fannie Mae. Hennessey, Holtz-Eakin and Thomas call the majority account too broad: "When everything is important, nothing is" (Financial Crisis Inquiry Commission, 2011, p. 414). Wallison objects that Pinto's research was never made available to the other commissioners (Financial Crisis Inquiry Commission, 2011, p. 448); I could not check whether the majority answered the point.

Acharya and colleagues. The 2011 paper is ADBI Working Paper 264, by four authors at NYU Stern School of Business; I read the working paper, not the book chapter it is to become. The authors state their own lens: many observers, "including ourselves", see the crisis "first and foremost as a regulatory failure" (Acharya et al., 2011, p. 1). The 2013 piece is Acharya's own summary of his research in the NBER Reporter, with no page numbers; I read all of it but none of the underlying papers in its footnotes.

Tarr. The library lists the journal version, Journal of Financial Economic Policy 2(2), pages 163–186. I could read only the World Bank Policy Research Working Paper 5324 (May 2010), so details may differ from the journal version, and I give no page numbers for Tarr and quote only his word "toxic". His claims on the GSEs and the CRA rest on other people's work (Pinto, Wallison, Hossain, Calomiris), which I have not read. The FCIC majority concludes the opposite, and Hennessey, Holtz-Eakin and Thomas side with the majority on the CRA.