The aftermath of the global financial crisis left many people with an impression of modern finance as extraordinarily complex, a web of multi-layered securities financed through strange off-balance sheet vehicles and held together by impenetrable mathematical formulas. There is something in this account but it’s important to remember that the most important reasons for bank failures are the old, traditional ones.
The easiest way to bankrupt a bank is to lend too much to people who are unable to pay you back.
The type of lending varies over the decades: excessive lending to railways, natural resources, emerging country sovereigns, commercial real estate, agricultural land and household real estate have all been the subject of bank failures over the last century or so. (The US Federal Deposit Insurance Corporation has a very helpful list of publications on the history of US bank failures which capture most of these examples, but the railway booms and busts in the UK and US preceded the deposit insurance era, which started in 1933).
The failures of Bear Stearns and Lehman (and Enron (*) earlier which, although an energy company, shared much of the complexity of an investment bank ) arose from a mixture of poor investment decisions (the asset side of the balance sheet) and fragile funding sources (the liability side). Many banks found during the crisis that they were unable to rely on their normal sources of funding, especially those that tapped the “shadow banking system“, meaning a range of short term funds provided through the commercial paper and repo markets. Banks traditionally funded themselves entirely from customer deposits and shareholder equity. But markets-based sources of funding became cheap and plentiful in the 1990s and 2000s as a result of financial innovation.
The modern form of bank runs
So banks came to rely on the markets to fund their assets and were shocked when those markets dried up or suddenly required far more onerous terms (higher rates and more collateral). A bank that suddenly cannot roll over (refinance) its funding is in the same situation as a bank that suffers a sudden loss of customer deposits, both are a form of bank run. If the bank has assets (its loans) which are still of good quality, then it is technically solvent and may be able to ride out the storm if it can get an alternative source of funding, perhaps from the central bank. According to an old doctrine, central banks should lend to solvent but temporarily illiquid banks, since otherwise the bank will needlessly fail and risk damaging other banks that are essentially sound.
But what we learned as the financial crisis developed was that most of the banks with funding troubles also had severely impaired assets, which is a fancy way of saying that they had made bad loans. A case which is pretty clear cut and now well documented is the British bank HBOS, whose name comes from the merger of the Halifax building society (a bank, originally mutually owned, which specialised in mortgage lending) and the Bank of Scotland.
The HBOS disaster
The Parliamentary Committee on Banking Standards was set up by both Houses of Parliament to look at the problems of the British banking system in the wake of the financial crisis, LIBOR fixing scandal and public concern about the state of one of the UK’s biggest industries. Their fourth report, published in April 2013, looked at the failure of HBOS, which was bought by Lloyds Banking Group, at the government’s behest, in January 2009. HBOS had a peak market value of £40 billion in 2007. Since then it has cost the UK taxpayer £20.5 billion in direct support to HBOS and indirect support to Lloyds, whose shareholders have also made multi-billion pound losses on the acquisition. The original HBOS shareholders lost 96% of their money and the 4% left is only there owing to the state-supported acquisition by Lloyds.
The Committee’s report is admirably clear, short and very informative. It is written in a tone of barely controlled anger because of the unwillingness of the management to acknowledge how disastrously they had run the bank. HBOS suffered from the drying up of wholesale funding markets that affected many other banks. But that merely brought to attention the huge losses the bank’s reckless lending expansion had brought.
£47bn in losses
The bank managed to lose money in three separate areas: £25bn in corporate lending; £7bn in its treasury operations; and £15bn in its international division. In Ireland for example (part of international) most banks made huge losses, but HBOS, which had an official target of becoming the number one business bank, experienced bad loans of 35%, second only to the truly catastrophic 48% of Anglo Irish Bank (which was nationalised by the Irish government in 2009). In Australia, the only major mature economy not to suffer a recession or banking crisis, HBOS managed to lose 28% of its loans. The corporate lending losses owed nothing to complex structured products or fancy securities. It was just the result of a policy of growing the bank’s scale and market share. If you give sales people incentives for selling, without penalty for the quality of the business they sell, then they will indeed sell. Loans are like any other product. And HBOS’s bankers went out and sold.
When interviewed by the Committee, the senior management team admitted things had gone wrong, but insisted that if the crisis in market funding had not happened then the bank would have survived. The fury of the Committee’s members at this is understandable. It is quite clear from the data that HBOS was insolvent (it’s tangible book value was £18bn in 2007). The funding crisis was just the manner in which the bad loans were exposed.
In fact the management’s claims are quite disturbing. The fundamental problem at HBOS was huge losses on its loan book, which is the asset side of the balance sheet. The problems of funding were on the liability side. Did the management really not understand the difference? The Committee’s report is carefully written but its scorn for the management team runs through the report like a scarlet ribbon of contempt.
The specific failures of management are mercilessly documented in the report, including a misaligned risk function, inadequate stress tests, board discussions that were too high level to actually challenge the operational decisions and a general willingness to hear only good news (as early as 2004 the board ignored the group finance director’s message that the bank’s rapid growth had created “an accident waiting to happen”).
There is a link between the liquidity and funding problems on the liability side and the poor assets. Most banks suffered some degree of funding and liquidity difficulties in 2008 but the market’s correct suspicion that HBOS’s assets were more at risk than those of other banks meant that it was worse hit. The conclusion is that the management were doubly wrong. Reliance on wholesale funding is dangerous, even for a bank with strong assets on its balance sheet. But if your assets are dodgy then relying on short term market funding sharply raises the risk that, when something bad happens, as it surely eventually will, you have no room for manoeuvre.
Note (*) Enron has a reputation for financial cleverness, which is partly deserved, owing to their pioneering of new areas of risk management and trading, including weather derivatives. But one of the causes of the company’s failure, the biggest bankruptcy in US history at that time (2001), was too much leverage in the company combined with poor investments, just like HBOS. A major error was the purchase of the British water company Wessex Water for a 30% premium in 1998. Enron then refloated a financially restructured version of the company on the New York Stock Exchange in 1999, renamed Azurix, presumably hoping that dozy American investors would not notice that it was pretty much the same British company with an exciting new name. Investors may sometimes be dozy but they are not completely stupid and the shares of Azurix (34% owned by Enron) fell by about 90% in the next two years, leaving Enron and the charismatically inept CEO of their international division Rebecca Mark in trouble, and thereby contributing to the company’s bankruptcy at the end of 2001.