Some time back, someone shared a new acronym with me: GFC. For those, like me, not in the know, this is apparently the hip shorthand for the global financial crisis – the economic mire that engulfs us all. It has been going on, more or less, for five years now, with every indication that it will get worse. It is its evolution from a banking crisis to a sovereign debt crisis – at least within the Eurozone – that gives me my optimistic outlook. I’ve been trying to write this post for ages, but every time I do, the story gets bigger and and seem to be constantly changing. It still is.
I have written this in an attempt to understand how we got here. I must stress that, whilst I have taken a couple of classes in economics, I am not an economist. I am sure I am missing large chunks, and I look forward to being corrected by people who have a greater understanding of these matters than I have.
Trying to understand, I’ve read three books on the subject in the last year or so – Gillian Tett’s “Fool’s Gold”, Michael Lewis’s “The Big Short“, and Andrew Ross Sorkin’s “Too Big Too Fail“. There’ve been innumerable TV programmes too – the most accessible exposition was Inside Job, shown as part of the BBC’s Storyville strand (and hence likely to be repeated).
(There are of course many, many books on the collapse of our economies: there is a veritable book bubble blooming.)
“Fool’s Gold” is probably the most useful source – Tett explains the acronyms – sorry, the derivatives – behind the crash in very clear, understandable terms: how and why they were invented, and how their abuse managed to more or less brought capitalism to a halt. “The Big Short” essentially looks at other side of the deal: traders who were selling (and insuring against) the toxic derivatives which the large banks were stockpiling. (A “short” is basically a sale of an asset, sometimes mediated by other derivatives – which means you can sell something you don’t own…) Michael Lewis’s heroes are all outsiders: they are people who looked at financial institutions’ assets, took a contrarian view and sought to profit from it. Some of them profited handsomely.
“Too Big Too Fail” is a big book; too big to read, perhaps… It is very US-focused and very detailed, describing who met whom and what they discussed. All the protagonists seem to share names, and they have all worked at each others banks, so it gets pretty confusing.
And this of course is a big part of the problem: there were huge conflicts of interest. But I’ll come back to that…
I want to distil what I think I learned from these and other sources – my own view of how we got here. It has grown into a long post. None of this is original, but here is what I think has happened…
Back in the 1980s and 1990s, banking was deregulated: Big Bang in the UK removed the distinction between retail, commercial and investment (popularly called “casino”) banks, and the repeal of the Glass-Steagall Act in the US did the same. At the time I thought this was a good idea – and perhaps it was; but it freed up banks to get much bigger – “too big to fail”, as Sorkin puts it – and to take each other over in new, novel (and risky) combinations.
Ironically, Glass-Steagall was introduced after the “Great Crash” of 1929, to stop a similar catastrophe from happening again. And it kind of worked, until they removed it.
As a result of deregulation, banks could use funds from customers’ deposits – cheap money – and lend it to corporate clients; and they could borrow money from others – other banks, or the money markets – to fund their own activities. They were allowed to trade on their own account, too. (Note again – conflict of interest…)
In both the UK and the US, deregulation was coupled with “light touch” regulation. The regulators – a whole plethora of them in the USA, headed by the Federal Reserve; in the UK, a flock of regulators were brought together to form the Financial Services Authority – assumed that it was in banks and other institutions’ interest to understand exactly what they were doing, and the stock and bond markets would know and understand what they were doing, and price their shares, bonds and assorted products accordingly. They put their faith in markets – the market would price all assets correctly.
This bred complacency: because it worked for so long, it must be working. Gordon Brown believed he had eradicated boom and bust. (The hubris in this belief is stunning: centuries of economic thinking overturned by one Chancellor of the Exchequer. Economic cycles are deeply engrained in our economic culture, possibly a throw-back to agriculturally dependent times; in industrial economies, boom and bust might be thought of as market failures arising from mispricing of assets and risk, first one way, and then another.)
In the UK, the home mortgage market was deregulated, so any bank could create mortgages. In the USA, the large, formerly state-owned mortgage lenders and guarantors “Fannie Mae” and “Freddie Mac” traded on an implicit state guarantee, lent recklessly – under political pressure, because a bouyant house market was deemed to be good politics.
Computing power allowed very bright people to create new financial derivatives. And they were very bright – a friend of mine who worked at Drexel for a short spell in the 1980s is now a professor at Oxford University. (Drexels went bust a few recessions ago.)
There is nothing new about derivatives: farmers and agricultural merchants have used futures (selling something now for delivery in the future) and options (selling the option to buy something at a set price at some point in the future) for centuries to manage their cashflow and balance risks: by using futures and options together, they could even insure against crop failure (sell a future and buy an option to purchase what you have sold in case the harvest fails). These are called derivatives because the are derived from underlying assets – you are not buying the asset itself (corn, or oranges, or – famously in “Trading Places” – pork bellies) but a bit of paper derived from the asset that gives you rights or obligations over the asset.
Futures and options – the bits of paper – could then be traded on financial markets. You take the bit of paper and sell it. One – key – point: you don’t need any interest in the underlying asset to buy or sell the derivative. I could go and buy pork belly futures if I could find someone to trade with.
The price of the derivative depends on the market price of the underlying asset – in ways I have no understanding of: the valuation of derivatives is determined using the Black-Sholes model which gained its eponymous modellers the Nobel prize in economics. (Like I said – these were bright guys – but it is interesting to note that Myron Scholes and his fellow Nobel laureate Robert Merton were both on the board of Long-Term Capital Management, a hedge fund which blew up in 2000 and was then the largest financial disaster. LTCM made huge bets on market movements funded by borrowings – that is, they were highly leveraged. Their models said they had made the right bets, but the market disagreed and they went spectacularly bust. LTCM may have been right – but as John Maynard Keynes said, “markets can remain irrational far longer than you or I can remain solvent” – a quotation we would do well to remember.)
With a way of valuing derivatives and cheap computing power, lots of different derivatives could be designed. For instance, if you want income (say you’re a pensioner) and I want capital growth (because I don’t want to pay income tax at the moment, perhaps), financial institutions designed investment trusts which split the capital growth of the underlying assets – shares – from the income stream they delivered, and lo, split capital investment trusts were created. (Many of these went bust in late 1990s and early 2000s – I’m assuming you can see a pattern emerging here…)
Lots of these derivatives involved “securitisation” – bundling up loans of one sort or another into a bond, and selling it on in the bond market. Securitisation is just a way of selling on an asset, really. The bit of paper – the security – is of course meant to be just that – secure.
The bright young things at J.P.Morgan came up with new securitised derivatives that got called colateralised debt obligations (CDOs). A financial institution – a bank, say – could bundle up mortgages made to borrowers (secured against their property) into a bond, and then sell it on – the mortgage loans have been securitised, allowing the bank to issue more mortgages because the bond – the CDO – has been sold on and is no longer on their books, and the bank has more money (from the sale of the CDO) to lend out.
Derivatives helped banks manage their balance sheet – essentially, they were selling on the risk of default. It also allowed them to leverage their lending: the bank could borrow money, lend it to housebuyers, bundle it up with loans made to lots of other homebuyers, sell it as a CDO and then lend the money generated by the sale of the CDO to housebuyers, and so on ad infinitum.Indeed, it has to lend it out to maintain its income from assets: a bank with too much money doing nothing will not be as efficient as one which has lent all it can.
The people who bought the CDOs – other banks, perhaps, or pension funds, or (apparently) German dentists – would get an income stream (the interest from each mortgage, paid on by the issuer of the bond) and, when the bond matured, repayment of capital.
A Bit About Portfolio Theory…
The people creating and selling CDOs came up with another whizzy idea. Financial markets had long subscribed to portfolio theory. This states (roughly) that combining assets with different risk characteristics can reduce the underlying risk of the portfolio to more or less zero if the price movements of the assets are not correlated. Only the specific risk of holding the asset can be diversified away – the systemic or market risk cannot be. For instance, a portfolio comprising coffee and wheat will be less risky than one comprised of either coffee or wheat alone, assuming that the market price of coffee is not correlated with wheat. This theory – a development of which also won its creators the Nobel prize – is commonly used to create investment portfolios.
One underlying assumption is that the prices of assets within a portfolio are not correlated. Another is that the prices of assets traded on a market are correctly priced by the market in the absence of any new information: all known information is incorporated into the price. This itself is based on the assumption that all investors are rational. The frequency of economic bubbles and the subsequent bust tells us that this assumption doesn’t hold – but it is a simplification to allow for easy modelling.
What the CDO makers did was bundle up homeloans that they believed would be uncorrelated, and thereby reduce the risk of holding the CDO. Then they bundled up different CDOs, and “diced and spliced” them so that investors –could choose the amount of risk they wanted to take, and the amount of income they wanted: these CDOs formed out of other CDOs are sometimes called “synthetic CDOs”.
Credit Rating Agencies
A lot of investors want to know the risk of the assets they hold – CDOs, in this case. Very often, if they are an institution like a pension, their corporate rules mean they can only hold the least risky assets – usually “triple AAA”. These are the most secure but pay the lowest level of interest.
They use information published by credit rating agencies – Standard & Poors, Moody’s, and Fitch – to ascertain the credit rating of an asset.
Because CDOs and synthetic CDOs were created using a portfolio to diversify away the risk, the credit rating agencies gave them high ratings – whatever the risk rating of the underlying assets. Because portfolio theory said the risk had been diversified away.
This was good for the banks creating CDOs, who could then sell the CDOs to say, German pension funds who could only by the assets with the lowest level of risk. The CDO creators included the credit rating agencies information in the CDOs’ documentation. And they paid the credit rating agencies for their work.
I think maybe I’ll repeat that: the people measuring the risk of CDOs were paid by the people trying to sell the CDOs to people who required highly rated assets. Credit rating agencies made a lot of money given opinions on CDOs. And the agencies were chosen by the banks creating and selling the CDOs – on whom they depended to get paid.
Thank God there is no conflict of interest there.
There were many other types of derivatives. Like mortgages backed CDOs, it is possible to produce new derivatives backed against any kind of an asset. A derivative called “credit default swaps” allow the holders of debt to insure against the risk of the default. This is a very useful tool – if you hold a bond from a company and you are worried they may go bust, you can buy a CDS to protect against that risk.
Another twist though: you don’t have to own the bond to buy a CDS insuring against its default. This is a bit like me buying insurance that will pay out if your house burns down. You wouldn’t even know I had done so, but if your house burns down, I make lots of money.
Institutions holding CDOs took out CDSs to insure against the risk of the CDO going bust. Because the credit ratings agencies reckoned this risk was very low – because the portfolio of mortgages in a CDO had diversified away the risk – these CDSs were very cheap.
A booming market in CDOs had created a booming market in CDSs, and the institutions writing the CDSs – banks and insurance companies, mostly, made a lot of money in fees. One of the main issuers of CDSs was the UK-based investment arm of an American insurer called AIG.
The size of the market was vast, because investors who thought CDOs were misvalued and could go bust bought cheap CDSs against them (this is “the Big Short” Michael Lewis writes about) or any other asset class. At the end of 2007, the notional market value of CDSs – the value that the insurance covered – was $62 trillion. (By comparison, the GDP of the USA in 2010 was $14.5 trillion: so even allowing for a fall in GDP as a result of the GFC, the market in CDSs was several times the whole economy of the USA.)
“Fool’s Gold” indeed.
The Value of Paper
All these derivatives – the CDOs, the CDSs and a plethora of others – were hard to value. There weren’t open markets – you could buy CDOs from the institution that issued them, but not sell them in the open market. Prices were calculated from the financial models used by the originating institution, based on the assumptions they used (such as portfolio theory).
It is fair to say that these models used all the available data to value the assets and calculate the various bits of risk they contained – similar models were used to value the CDSs and the CDOs they insured.
It is also fair to say that the price history of these derivatives didn’t go back very far. The people who created the models were generally young, who couldn’t remember back to – any financial crisis, really – not the stagflation of the mid 1970s, and definitely not the Great Crash of 1929. So while their models used all the available data, that isn’t necessarily saying much. To make up for their lack of data, they extrapolated from the data they did have.
Maybe not so bright.
The Dotcom Bust and 9/11
(Yes – I believe that Osama bin Laden was at least partly responsible for bringing western economies to their knees.)
In 2000, a stock market bubble built up of investments in (then new) dotcom internet companies burst as a result of increasing interest rates.
In the response to the attacks, the Federal Reserve dramatically reduced interest rates to ensure liquidity didn’t collapse resulting in financial crisis. Over three months or so, the interest rate halved – from 3.5% to 1.75% [pdf].
Money became very cheap.
This lead to a stockmarket boom – and to people chasing income: because interest rates were low, people looked to other assets which could generate rates of income greater than bank interest.
Assets like CDOs.
Housing Markets Boom
Low interest rates also lead to a boom – a bubble, even – in house prices around the world: low interest rates meant cheap mortgages, so lots of people could afford to buy their homes.
This meant more mortgages – and more CDOs.
There was another twist, though. In the USA as well as many other places, mortgages were sold through mortgage brokers. So the banks were middle men – brokers sold mortgages so their clients could buy houses, and the banks sold on the mortgages as CDOs. The brokers completed the forms, and the banks sold on the loans as CDOs to other investors, churning the mortgage and then relending the proceeds of the sale of the CDOs. The banks therefore had no ongoing relationship with the homeowner – the borrower. The brokers had no relationship with the borrowers either: once the paperwork was signed (and the broker had earned their commission), they were away.
And the CDO investors had no relationship either – indeed, they probably couldn’t identify which mortgages (and the property they were secured on) made up their CDO, because after all that slicing and dicing and reconfiguring into synthetic CDOs – well, it would be anyone’s guess. The links between different parts of the chain were broken.
Worse, the banks and the brokers wanted the worst types of mortgages the most. Creating CDOs that paid high interest for low risk (because the risk had been diversified away) meant that they sought out those mortgages that paid most interest. These are the “sub-prime mortgages” were talked about so much a few years ago. They are sub-prime because individually they are risky – and so carry high interest. Once you bundled lots of sub-prime mortgages into a CDO, though, they became low risk – so said portfolio theory, at least, and that’s what the credit rating agencies said, too.
Investors were seeking income because of low interest rates; CDOs based on sub-prime mortgages offered high interest with low risk. So banks and mortgage brokers actively sought out the worst types of mortgages to bundle up and sell, in a game of rather toxic pass-the-parcel.
The flipside to the low interest rates was that it was cheap to borrow. So lots of people borrowed a lot of money – home-owners borrowed against the value of their homes (including those sub-prime borrowers at the bottom of the pyramid), people borrowed on their credit cards, local and national governments borrowed. (It would appear that, unlike everybody else, corporations didn’t borrow excessively.) Banks were throwing money at people, and the stock of debt went up and up. This money was spent on assets and goods, so asset prices went up – reducing the yield, causing people to seek higher income in riskier assets, feeding the bubble. (And repeat.)
Nothing could go wrong now.
So far, nothing is this story has been illegal. No one has broken the law. They may have been naieve, or so clever they were stupid – hindsight being a wonderful thing – but nothing criminal had happened.
Some mortgage brokers or some of their clients (or, most likely, both) did however do something illegal. They sold mortgages to people they shouldn’t have done, and the borrowers were complicit. These mortgages were mainly “self certification” mortgages, and the brokers (hungry for their commission) happily signed off credit checks they shouldn’t have done. They also sold people products they didn’t understand.
But since the house market was booming, it didn’t matter, because the mortgages were secured against ever-increasing property values.
Off-Balance Sheet Vehicles
Actually, it is possible there was other criminal activity. Lots of the derivatives – CDOs, mostly – were marketed as if they were stand-alone companies, and they were set up that way: a company would be formed to hold the derivative, and then bits of the company would be sold to investors. This way, none of the assets or liabilities would show up in the originating institutions books: they would be “off-balance sheet”.
This may not have been strictly illegal – the lawyers and accountants and financial whizz-kids who set these “special purpose vehicles” (SPVs) would probably make sure they were completely legal, in fact. But use of SPVs meant that no one looking at the accounts of the institution would have any idea of their involvement in the scheme. This should perhaps have flagged up some issues, though: SPVs (or similar off-balance sheet vehicles) were behind the big corporate failures of Enron.
Bank financial statements are hard enough to read at the best of times (I have tried, and I’ve had training to do so!); moving assets and liabilities into SPVs made it impossible to know quite where the buck would stop, and who owned what.
Which isn’t a problem when everything works as it should.
What Went Wrong
Interest rates went up as fears of inflation grew. The prime rate – the rate US banks charge their best customers – started to increase in 2004, rising a whole percentage point from 4% to 5% between June and December 2004 (that is, the amount of interest you paid on your variable rate mortgage increased by 25%), reaching 7% in November 2005 and 8.25% in June 2006 (where it stayed until September 2007). The UK base rate similarly stood at 4% in February 2004, rose to 5% by November 2006 and reached 5.75% in July 2007 – not such a steep rise, but significant nonetheless.
If you took out a mortgage in early 2004 in the US, the rate of interest more than doubled in two years. People who had taken out cheap mortgages found that they were now very expensive. People with the worst credit and security over their property – those with sub-prime mortgages – were worst hit by rising interest rates.
They started to default on their payments, and the housing bubble burst: as more people defaulted, the price of homes in the US fell until they were not sufficient security for the mortgage on the property. The value of property continued to fall – and so did the value of CDOs based on mortgages on the property.
All those mortgages which the modellers had assumed to be uncorrelated when they created the synthetic CDOs turned out to be highly correlated indeed: the USA housing boom bust everywhere and at all levels, and affected the bottom end of the market – those subprime loans – worst.
In June 2007, hedge funds with exposure to sup-prime mortgages and CDOs started reporting significant losses. Because they were heavily leveraged – hedge funds could borrow to increase their exposure to assets – a relatively small fall in asset value could have a calamatous effect on the capital of the fund. Investors panicked and tried to pull their money from the funds. But since the asset valuations were based on models which assumed that there would not be significant mortgage defaults (because they hadn’t been seen for several decades – well outside the scope of the models), no one knew the real value (or otherwise) of the funds and the assets they contained.
What trading there had been in CDOs and CDSs dried up: they became illiquid investments. All those financial modellers hadn’t bargained for this – an extreme event: maybe a once-in-a-hundred years event. Their valuation models hadn’t bargained for something akin to the 1929 crash.
Markets Dry Up and the Bubble Bursts
There was a knock on effect. Bear Sterns, a Wall Street stock broker, was associated with some hedge funds focussing on subprime mortgages (albeit that the funds were set up to be off balance sheet). Other banks – particularly JP Morgan, according to Tett – pulled in their loans to Bear Sterns. No one wanted to be left holding the baby. Since Bear Sterns couldn’t get credit, other lenders got more worried and withdrew their financing, so Bears Sterns lost more loans and had to be rescued.
Financial markets being very illiquid: not knowing which other banks might be in danger (if only because no one knew where all the toxic assets had ended up, what with all that slicing and dicing), banks ceased to lend to each other: this was the start of the “credit crunch”.
In the UK, Northern Rock, a small bank which had based its business model on making loans, securitising them, selling them on and borrowing in the market to finance more loans, found itself unable to refinance its loans. It collapsed and had to be rescued by the British government.
After this, the markets looked for the next bank likely to collapse – and, as “Too Big To Fail” describes in detail, eyes alighted on Lehman. The market was famously proved right.
Because no one could identify where the bad debts – the rotten CDOs – had ended up, because of all that slicing and dicing – no one knew which banks were holding it – and hence which banks were bad investments. Interbank lending completely dried up. Central banks – the Fed, the Bank of England and the EuropeanCentral Bank – cut their interest rates dramatically again to try to boost liquidity. (Perhaps pointlessly – banks weren’t lending because they were scared that the other banks they were lending to would go bust, not because interest rates were wrong but because they were scared. Indeed, since low interest rates had contributed to the race to the bottom by income-seekers, keeping low interest rates could just be building up the next bubble.)
Banks weren’t lending to each other, and many of their assets – all those subprime mortgages, CDOs and anything derived derived from them – became unsellable. And hence worthless, hitting banks’ balance sheets – which they desperately sought to shore up with fresh capital. Some took it from the US government through the “Troubled Asset Relief Program”, from sovereign debt funds (which ties into the LIBOR scandal – fiddling LIBOR may have helped Barclays look like good value to its new investors), other banks (such as the Lloyds takeover of HBoS at the behest of the UK government) or the UK government bailout package. Essentially, lots of big banks became insolvent and, one way or another, were rescued.
“Too Big To Fail” tells a great story that the US regulators and government blamed the Bank of England and Alastair Darling, the then Chancellor of the Exchequer, for blocking the takeover of Lehman by Barclays – though they didn’t ask till late on Sunday night and needed a decision by the opening of markets on Monday. So Darling said no. Barclays subsequently bought the bits of Lehman it wanted after the bank went bust.
Without liquidity and capital, bank lending dried up, pushing the economy into recession. People got worried and stopped spending. Though official interest rates were low, banks weren’t lending and no one was borrowing. Western economies more or less ground to a halt.
Lenders had to reassess the riskiness of assets they held, and markets reflected this riskiness. The market prices for some government debt fell drastically as investors worried that perhaps governments may not be able to pay back all that cheap debt they had earlier acquired.
In the Eurozone lenders had previsously assumed that any government debt was as risky as that of the least risky borrower – Germany. As debt markets dried up, it became apparent that that wasn’t a safe assumption. Interest rates for Spain, Ireland (both of which had undergone economic booms, fuelled by banks borrowing cheaply in the markets and lending to property speculators), Portugal and Greece (the latter of which had huge levels of government debt) saw the interest rates they were effectively paying on their debt rise sharply, to unsustainable levels: they would not be able to afford to refinance their debts as they fell due.
Throughout Europe and much of the rest of the world, governments have been trying to reverse the burden of debt and stabilise their finances – through austerity measures (largely public sector cuts) and, sometimes, tax rises. As many commentators have pointed out, these can have opposite effects: austerity can reduce the tax take and increase the demands on benefits.
Parties of the left condemn public sector cuts; parties of the right focus on austerity. Both seem to believe that economic growth is the answer to the world’s economic woes.
And this is roughly where we are now.
There Is No Answer
To me, the focus on growth seems to be missing the point: it was a focus on untrammelled growth since the Second World War that has landed us in this mess. To generate growth, central banks have cut interest rates to very low rates – even down to negative real rates (after the effect of inflation). It was low interest rates that fuelled the financial bubbles of the early 21st century – the dotcom boom and the US housing market.
Surely we are in danger of endlessly repeating the same diagnosis and taking the medicine that has failed before?
My uncomfortable conclusion is that something more radical is needed: a move away from a focus on ever-more growth to a more sustainable economy.
I have no idea what this might be, nor how it might come about, and somewhat pessimistically I doubt it will happen.
Clearly, great regulation is needed, and the multiple conflicts of interest inherent within our financial system need to be managed if they can’t be eradicated. What else might be needed or could happen reamins to be seen.
But if anyone claims that we have moved to a new world where the old economics have no meaning – as they did during the dotcom bubble – or that they have erdaicated boom and bust – well, you know that is your sell signal.