Read the article in the Cayman Financial Review Magazine
The operatic scale of the financial crisis that has engulfed the world since 2007 has finally been recognised. April 7 sees the premiere, in Talinn, of Nostra Culpa, a 16 minute opera inspired by a “Twitter feud” in which Estonian President Toomas Hendrik Ilves denounced New York Times pontificator Paul Krugman for asserting that Estonia’s response to the crisis had not been a success.
The feud highlights the contrast between those who believe the crisis justified massive government intervention and those who believe that such intervention is counterproductive. Recent research suggests that governments should cut spending during crises, not increase it.
On 6 June, 2012, Paul Krugman blogged:
“Since Estonia has suddenly become the poster child for austerity defenders – they’re on the euro and they’re booming! I thought it might be useful to have a picture of what we’re talking about. So, a terrible – Depression-level – slump, followed by a significant but
still incomplete recovery. Better than no recovery at all, obviously –
but this is what passes for economic triumph?”
See Figure 1
In response, Ilves Tweeted: “Let’s write about something we know nothing about & be smug, overbearing & patronizing: after all, they’re just wogs.”
So, who is correct? Let’s look at the arguments and the evidence.
Financial crises are characterised by severe reductions in asset prices and consequent collapse in fixed investment, as individuals and businesses seek to mend their balance sheets. Krugman and other interventionists claim that this fall in fixed investment by the private sector necessitates an increase in investment by the public sector. Without this compensating increase in public sector investment, say the Krugmanites, economic output would fall dramatically, risking a 1930s style depression.
Opponents say that this ignores the underlying problem that caused the recession in the first place, namely mal-investment. A financial bubble occurs when there is overinvestment in one or more sector(s) of the economy.
Put simply, the average returns from assets in those sectors are lower than is implied by the prices of those assets.
At some point, enough investors realise that the assets are overpriced and the bubble bursts. Often that is the end of the story and everything returns more or less to normal. The dot com bubble and bust is a good example: while some people who had made investments with borrowed money suffered personally when the bubble burst, the problem was not sufficiently widespread to impact bank lending.
But when the bubbly asset is something that most households own and have purchased by borrowing money, the problems can be much more severe. That was the case in the 2007 bubble, which began in the US housing market. Prices in that market were driven up by an overabundance of credit caused mainly by three factors: first, the under-pricing of debt by the Federal Reserve; second, the securitisation of over a trillion dollars in subprime and Alt-A mortgages by government sponsored enterprises (chiefly Fannie Mae and Freddie Mac); third, differential regulatory requirements for financial organisations holding mortgages as compared to mortgage backed securities (MBSs), which incentivised the creation of the latter and their use as capital.
To make matters worse, other central banks, eager to prevent their currencies from appreciating, engaged in competitive under-pricing of debt. This helped enable the bubble to spread well beyond US housing. Indeed, both Spain and Ireland experienced property price explosions at the same time. So, when the property bubble burst in the US, it spread rapidly to other assets and thence more deeply across the global economy, as major financial organizations holding MBSs as capital approached insolvency.
In August 2007, Northern Rock – then Britain’s fifth largest mortgage lender – found itself suddenly unable to roll over its short-term debt because the markets for MBSs were essentially frozen, so it went to the Bank of England, which extended a line of credit. In 2008, Royal Bank of Scotland, which had just purchased the US assets of ABN Amro for about £10bn, saw the value of its stake turn negative – and with that the solvency of the entire bank was put into jeopardy. In 2009, RBS wrote down its ABN Amro stake by £20bn.
Meanwhile, the popping of the property bubbles in Spain and Ireland had similar consequences to the property collapse in the US: many of Spain’s cajas (savings banks – and the main suppliers of mortgages) became insolvent; the Bank of Ireland and Allied Irish Bank (two of the country’s largest) came close to insolvency.
The governments of nearly every country in crisis then did pretty much what the Krugmanians proposed. The UK government was a first mover, taking over Northern Rock in February 2008. In March 2008, the New York Fed bailed out Bear Stearns through a $29bn loan to a shell company, Maiden Lane LLC, run at arms’ length by JP Morgan Chase, which also loaned Maiden Lane $1bn.
In September 2008, the New York Fed offered AIG an $85 billion credit facility in return for warrants for 79.9 per cent of the company’s equity. In the same month the Irish government announced it would guarantee all bank deposits up to €100,000. In October, the UK government partly nationalised RBS and Lloyds-TSB, sinking £37 billion of taxpayers money into the two banks.
Those hors d’oevres were followed by the $700 billion Troubled Asset Relief Program, which arbitrarily imposed government debt on a wide range of US financial companies – and a similar programme in the UK. The Irish government intervention of February 2009, in which it “rescued” AIB and Bank of Ireland with a €7 billion package, looks modest by comparison – but that was just the beginning of the bailout. In December 2010, the Bank of Spain engineered the merger of seven failing cajas into Bankia.
Two years later, after becoming the third largest lender – and the largest holder of real estate assets – in Spain, the by-then insolvent Bankia was nationalised and received a bailout of around €20 billion.
While Krugmaniacs tout these interventions as successes – we are told that averted a global economic meltdown – they don’t usually discuss the downsides. For example, the massive expansion in government debt, as private debts were absorbed by taxpayers. Ireland’s government debt-to-GDP ratio went from 25 per cent in 2008 to 117 per cent today; the UK went from 44 per cent to 90 per cent; the US from 65 per cent to 102 per cent; and Spain from 36 per cent to 88 per cent.
In the years since the crisis, the economies of several other European countries have begun to look increasingly shaky. Portugal, Italy and Greece, in particular, have fared poorly – though their problems have more to do with longstanding dysfunctional government, endemic corruption and generally poor institutions.
Unlike the UK, Ireland and Spain, whose government debt-to-GDP ratios had been falling prior to the financial crisis, Portugal’s had been rising; going from 56 per cent in 2004 to 68 per cent in 2008. It currently stands at 120 per cent. Meanwhile, Italy’s debt-to-GDP ratio has been over 100 per cent since the early 2006 at least and currently stands at 127 per cent, which is worse than in 2008, when it was about 104 per cent, but not much worse. Greece’s official debt-to-GDP ratio broke the 100 per cent barrier in 2006 and rose to a staggering 170 per cent in 2012 before falling back to around 160 per cent.
But so what if governments have high debt-to-GDP ratios? After all, “debt is (mostly) money we owe to ourselves,” isn’t it? Well, Mr Krugman (who wrote those words), in a word: no!
Debt is money that one individual or group of people owes to another. If the debt holder is forced to take a haircut, it has implications for all manner of downstream transactions. Recall the above sequence of events: it was the haircut holders of mortgage debt were forced to take following the decline in property prices that resulted in a seizure to the credit markets, which in turn led to a further decline in asset prices across classes, which led to the collapse in investment that government intervention (at least in the US, UK, Ireland and Spain) is supposedly correcting.
It seems bizarre, therefore, to argue that haircuts don’t matter, on government or any other kind of debt. Beyond that, when governments threaten to default on their debt, yields quickly rise, often resulting in a vicious circle of escalating debt service costs and rising debt – until the inevitable default materializes and borrowing becomes more or less impossible.
Krugman subsequently argued that his main point was not that debt holders would have to take a haircut per se but that the debt would enable the government to stimulate growth, so that over time the debt could be paid off or would become inconsequential. Well, that at least gives us a testable hypothesis. Do government bailouts result in more rapid growth?
Since it is more or less impossible to increase government revenue during a financial crisis, the funds used to pay for the bailout must come from the sale of bonds. In other words, as we have seen, private debt is converted into government debt and many bad investments that would ultimately have been written off had they been left in the private sector become the responsibility of taxpayers. So, the next question is: does this increase in government debt result in an increase or decrease in economic growth?
In a paper published in September last year, Harvard economists Carmen Reinhart, Vincent Reinhart and Kenneth Rogoff analysed the impact of significant increases in public debt since 1800. They focused on periods – such as the present – when average public debt had risen above 90 per cent of GDP on a “sustained basis” and concluded that such periods are associated with lower rates of economic growth over a lengthy period (of the 26 episodes they identified, 20 lasted more than a decade – and the average was 23 years).
I think from this we can fairly conclude that bailouts that result in substantial increases in debt are likely to reduce economic growth, not increase it.
Part of the explanation for this is that when governments intervene to bail out banks and other financial companies, they effectively subsidize the mal-investments that led to the crisis (since they are not fully written down, as they should be) and lock the economy into the lower rates of innovation and growth implicit therein.
In a paper published in 2007, an economist at Munich University, Sebastian Haubtmeier, and two economists at the European Central Bank, Martin Heipertz and Ludger Schuknecht showed that countries that reduced government debt-to-GDP ratios over the past two decades experienced significant increases in rates of economic growth.
Meanwhile, a paper in the CATO Journal by Steve Gjerstad, an economist at Chapman University, showed that three countries that cut government spending following a financial crisis (Finland, Thailand and Iceland) experienced a more rapid return to growth, in part due to currency depreciation and an associated export boom that helped mend balance sheets rapidly.
So, what about Estonia? Well, the bad news is that its government debt-to- GDP ratio has increased significantly since the financial crisis. But the good news is that even so its debt-to-GDP ratio is currently only 10.5 per cent, suggesting that its prospects for continued economic growth are good. By cutting spending, the government managed to limit the impact of reduced tax receipts.
And of course it avoided engaging in any bailout shenanigans. Indeed, its growth rate has increased since dipping last year and is currently estimated at approximately 3.7 per cent. So, we are living in operatic times but it looks like the Krugmanloons were wrong and President Ilves was right.