Friday, 13 December 2013

The UK recovery and the pessimists’ refrain

In two recent posts I plotted UK GDP per person since 1950. What is remarkable to me about that time series is how well a simple trend tracks the data – until the current recession. In reality trend growth rates have probably moved around a bit, but the important point is that past recessions have essentially turned out to be temporary deviations from trend growth, rather than signaling fundamental shifts. In particular, following both the major recessions of the early 1980s and 1990s, we achieved recoveries that brought us back to something close to the level of output we could have achieved if there had been no recession.

There are two arguments that this time will be different: what I will call the pessimists’ refrain. The first is that we were fooling ourselves before the recession, because in reality we were ‘living beyond our means’. This argument suggests that output in 2007 was unsustainably high, so our trend line should be lower and flatter. The second is that since 2008/9 productivity has stalled because the financial system in the UK has been broken. I call these ‘refrains’, because they usually come with a repeated message: we should stop stimulating the economy, because if we try to get back the ground we lost, we will fail and instead generate inflation. (A third argument focuses on hysteresis effects of high unemployment, but this is more important for the US than the UK.)

Larry Summers has recently reminded us that there was something odd about the pre-recession period in a number of countries including the US and UK. Although inflation was reasonably stable, we had the kind of housing bubbles that we normally associate with booms. Now Summers argued that this indicated an underlying demand weakness: we could only get people to buy all the stuff we could produce by encouraging them to take out an unusual amount of debt. There is an alternative explanation, which is that - despite appearances - there was a boom before the recession. We were, in 2007, living beyond our means. 

This idea is not the view of a small minority. Organisations like the OECD and IMF now calculate that in 2007 output gaps were large and positive. The latest OECD Economic Outlook gives 3.3% for the OECD area as a whole, 3.5% for the Euro area, 2.9% for the US and 4.4% for the UK. That is not what these organisations were saying at the time. In the June 2008 Economic Outlook, the equivalent numbers were 0.4%, 0.0%, 0.4% and 0.2%. At the time it looked like output in 2007 was close to the natural rate in many countries, including the UK. [1]

This change of view on output gaps, where 2007 goes from balance to a significant boom, is largely inevitable given the way the OECD and IMF calculate these numbers. Since the recession productivity in many countries has been much lower than we might have expected (in the UK it’s the ‘productivity puzzle’), which seems to indicate a fall in how much technical progress has been embodied in production. Traditionally we have thought of technical progress changing gradually, and as being largely unrelated to the economic cycle. If productivity is low today and technical progress only changes gradually, then it follows that some of this slackening in the pace of technical progress must have started before the recession. So in 2007 the economy was not capable of producing as much as we thought at the time.

What hard evidence do we have for all this? The living beyond our means case really comes down to the idea that both the housing bubble and the build up of personal debt must imply there was a boom. Yet this is really one piece of evidence rather than two. As Ben Broadbent shows, the build-up of debt was matched by an increase in assets: the value of houses. (The same appears to be true in the US.) This increase in personal sector indebtedness might have been foolish given what happened to house prices, but the underlying problem was the housing bubble.

Was the housing bubble an indication of a ‘hidden boom’ in 2007? There is a quite plausible alternative explanation, which is that you can get housing booms when real interest rates are low. Many economists, including Ben Bernanke, pointed out before the recession the unusually low returns on long term assets like government debt, an idea that became known as the ‘savings glut’. The return on assets was being driven down because consumers in China and its neighbours were saving extraordinary amounts, leading to large current account surpluses there. When returns on safe assets like government debt are low, people look elsewhere for higher returns, and this includes the housing market. This is probably why we are seeing rapid increases in house prices today in a number of major cities (e.g. London, Paris, Germany).

Not only is there this alternative explanation for the housing bubble, but the living beyond our means case cannot satisfactorily account for why - if we had a huge boom - inflation remained so subdued. What upward movement there was could easily be explained by large increases in oil and commodity prices. It is sometimes argued that inflation targeting, or cheap goods from China, kept a lid on consumer price inflation, but there was no indication of any overheating in the labour market either.

The second refrain, much more specific to the UK, focuses on the banking sector. The argument is not that the financial sector grew too rapidly before the recession, and must inevitably shrink back to a more normal size. That is almost certainly true, but the numbers just do not seem large enough. As Martin Wolf points out, the financial sector went from 6% of GDP in 1998 to 9% in 2008. Even if it returns to 6%, for the economy as a whole much of that should be recoverable, because most of those who have lost their jobs in the financial sector are highly skilled and so can be redeployed in other high productivity activities.

The argument is instead that, for the supply side of the economy to grow, more productive firms need to replace inefficient competitors, and new innovative start-ups should challenge existing firms. Both processes almost certainly require borrowing, and in the UK in particular that borrowing usually comes from a few large banks. If these banks stop lending, productivity growth will fall away. This argument is plausible, and can help explain two puzzles about the current recession. The first is why, when UK firms are asked how much spare capacity they have, they respond that they have very little. This is not consistent with the recession being only about lack of demand. The second is that inflation has not been falling more rapidly. If firms cannot get the finance to grow, there is little point in trying to expand your market by cutting prices.

What is frustrating about this idea is that there is little hard evidence either way. Bank lending to firms has certainly fallen, but how much of this is down to banks, and how much is simply that firms think it is too risky to borrow? (This study suggests at least some of the former.) There is a great deal of anecdotal evidence that some banks have in the last few years might have been hindering rather than helping small businesses, but translating this into actual numbers for productivity lost is almost impossible. The most compelling argument that something like this has been going on is that other explanations for the UK’s productivity puzzle are either implausible or inadequate in terms of scale.

Perhaps the key question, though, is how permanent this all is? If bank lending starts to recover, can we get back the productivity we have lost? We can look at past financial crises in other countries, as Nick Oulton has done. (His paper also provides useful detail on why other explanations for the productivity puzzle do not seem to work. The section on fiscal policy, however, departs from his usual high standards, as this article in Pieria suggests.) His analysis indicates some permanent hit to GDP from a financial crisis, but the larger numbers come from Latin America. The example of Sweden in the early 1990s is much more optimistic. A priori we might expect a good deal of the innovation to have been ‘put on hold’ because of lack of finance, and this could be activated once banks’ balance sheets are repaired. But perhaps some opportunities may have been lost forever.

Given these uncertainties, two implications for policy seem clear. First, we should stimulate demand until there are clear signs of overheating in both the goods and labour markets. We should only do otherwise if the pessimistic case is compelling, and it is not.

Second, to the extent that we do not recover the ground that we lost in the recession, the costs of the financial crisis are even larger than we thought. This suggests we must do something to make the economy less dependent on the behaviour of a small number of large UK banks. What is interesting about at least some of those who sing the pessimists’ refrain is that they seem to treat this implied loss of UK capacity like an act of God: not only is it something we can do nothing to reverse, but there is little point in investigating it further. That is a strange attitude to take, particularly given that there has been no equivalent productivity puzzle in the country where the banking crisis really started. (This may be another reason to praise small US banks: see Felix Salmon here.) Some who are pessimistic preach the usual neoliberal message for enhancing growth, including lower taxes on high incomes, but seem strangely uninterested in what has caused this alleged huge reduction in supply. Perhaps they fear the answers to the UK’s productivity puzzle will not be to their liking.

[1] For the CBO’s assessment of US potential, see Menzie Chinn.           


  1. Simon,

    I have to take issue with you about small banks. There is absolutely zero evidence that a very large number of small banks is better for the economy than a smaller number of larger banks, and considerable evidence to the contrary: the historical record of financial stability in the US is awful compared to countries with more integrated banking systems, including the UK. I wrote about this here (in response to Salmon's article, with which I almost completely disagree):

    Small banks in the US are most active in property lending, not business lending; even small corporates rely far more on the capital markets for financing. The various forms of QE that the Fed has undertaken have therefore benefited a far wider range of US corporates than the equivalent in the UK. Adam Posen has written about this. I would also guess that the fact that US banks don't keep most of their loans on their books enabled many of them to recover faster than UK banks, which were left with significant portfolios of impaired loans that have taken several years to unwind. This has nothing to do with the relative size of UK versus US banks, and everything to do with the fact that the US banking system is an originate-to-distribute model, whereas the UK system is originate-and-hold.

    However, that aside, I frankly think you are looking at completely the wrong cause for the difference between the US and the UK recoveries. The UK embarked on premature and ill-considered fiscal consolidation at the end of 2010 in conjunction with tightening of monetary policy (ending of the SLS). At the same time there were sizeable rises in oil prices that fed through into double-digit inflation in energy & fuel prices throughout 2011 and into 2012, which neither the Government nor the Bank of England did anything to counteract - in fact the Government's tax rises (VAT and green levies) on energy bills made the impact worse. The UK has also been far more badly affected by the Eurozone crisis.

    The UK has also tightened macroprudential regulation for banks considerably - probably more than any other country. This has a directly contractionary effect, since higher capital and liquidity requirements discourage lending, particularly to higher risks (such as SMEs).

    1. Frances

      I think this is very interesting so I’d like to pursue it. Of course I agree with your penultimate paragraph, but I’m not sure that can explain the productivity puzzle we observe in the UK: an apparent reduction in supply. This is why I talk a lot about banks – although, as I say in the piece, this is mainly because of the lack of other plausible supply side explanations.

      I suspect my real concern is not about size per se, as about competition and focus. By focus I mean that large UK banks are heavily involved in other activities besides lending to firms or households. When these activities go bad, the whole business suffers. Ultimately what we want is a financial system that is more robust, and instinctively I feel that means having more diversity and competition in the sector.

      But if it’s not the structure of the UK banking system that is the problem, what else is it that seems to have made the UK especially vulnerable to a reduction in bank lending? Your final paragraph suggests that it is tighter macroprudential regulation in the UK. The Bank of England strongly denies that their attempts to improve the long term robustness of UK banks is causing these banks to restrict their current lending – do you think this is wrong? Is there any evidence on this?

    2. Simon,

      I'm afraid I do think that tightening macroprudential regulation discourages lending to businesses, and I think there is substantial evidence for this.

      I don't think people realise just how badly damaged the UK banking sector is, though. It's far worse than just a small number of large banks. Every bank of any size in the UK is carrying substantial bad debts, and the building society sector is also badly damaged.

      I would say the problem in the UK (relative to the US) is the reliance of business on banks for finance. We don't have sufficient diversification in SME finance. And SME finance is the highest risk form of lending. Consequently, when the banking sector is damaged and/or regulation is tightened, SME finance is the first to suffer. If we had more non-bank sources of finance for SMEs we would not be so vulnerable. Have you read the Breedon report into non-bank business lending? Also, I would like to see more equity financing schemes for SMEs.

    3. Oh, and in the UK it was not "other activities" that went bad, except in RBS. It was lending to firms (especially commercial property) and households. That's why the UK's financial crisis was not limited to investment banks - indeed investment banking was not much involved, really. Fundamentally, the UK's financial crisis was a crisis of retail lending. And that, in a nutshell, is why we now have much less retail lending. Banks lent far too much at too high risk before the crisis: they have had their fingers very badly burned and are now lending not enough at too low risk (very high haircuts on collateral, for example). I can't say I blame them, but it's bad news for the economy.

    4. I don’t care how much we tighten up on bank regulation and how much the consequent reduction in lending is. That regulation should be tightened to the point where there is no subsidy (TBTF or other) of banks. Banks are supposed to pay their way. I fully back to call by Martin Wolf and Anat Admati for bank capital to be quadrupled to about 20% of bank assets or liabilities.

      As to the reduced AD that results from that tightening, that’s easily countered by monetary and/or fiscal stimulus.

  2. Simon, I'm sorry but I have a second disagreement too.

    I recently chaired a roundtable debate for ACCA SME in which representatives from Experian, Red Flag Alert and two insolvency practitioners discussed whether or not zombie companies were a major cause of the UK's poor recovery. Experian and Ernst & Young both presented compelling arguments that the "zombie companies" argument simply does not fit the facts. The other two presenters accepted that there might be zombie companies, but did not think that they were the most significant cause of poor recovery. What has changed is the nature of business in the UK: far more service companies and far more microbusinesses and sole traders, which tend to wind themselves up rather than going insolvent. And insolvency practice itself has changed, too. I used the key points from Experian's presentation in my post about the ASI's paper on zombie companies:

    ACCA SME are producing a video of the debate itself which will be available shortly.

    Experian said they feel very strongly that the "zombie company" myth must be dismissed. Their research does not support it and they are concerned that it may lead to inappropriate and damaging policy decisions. Max Firth of Experian wrote about this in the Telegraph:

    Experian don't exactly have any axes to grind, so I would take them seriously.

  3. I am puzzled by a couple of things here. One is the definition of "output gap". Is it the gap between current GDP and the GDP that could be produced by putting every piece of production equipment and every marginal worker to work, or is it the gap between today's GDP and the largest non-inflationary GDP that could be achieved. Two different things.

    Also, yes of course this recession was different to others. Post-war recessions in the UK have been Central Bank recessions, where interest rates are raised to head off inflation, but the most recent one was more like a Minsky debt deflation recession.

    Debt deflation recessions are always long, because they are caused by consumers and others de-leveraging. In other words, demand isn't just temporarily suppressed by rising interest rate; it really goes away.

    If such a recession is protracted enough, long term capacity contracts. Some production machinery goes out of use, skilled teams are dispersed, experience gets rusty.

    If that's the case, our estimate of output gap is just a guess. You said as much earlier this week when you displayed the graph of UK GDP from the IFS. It drew a trend line through several Central Bank recessions and concluded that there was something especially egregious about the current state of play. There is. The current state of play is a different enough recession that we should not expect previous trend-lines to be much help.

  4. You dismissively mention those who make the usual argument that the secret to more rapid growth is " lower taxes on high incomes," I strongly agree that you are right to dismiss them. There is less than no evidence that taxes on high incomes lower than current UK taxes would cause more rapid growth .

    On the actual topic of the post, I note that slack demand causes low productivity through labour hoarding. I have even less respect for standard estimates of output gaps than you do. I guess I am convinced by exactly one data point (and the usual one). US GDP returned to trend after the great depression and world war II.

    Well there is also the case of the UK in the 80s. In 1985 the UK was exhibit number 1 for hysteresis. The severe recession was followed by growth at around the normal trend rate with high unemployment. It was exactly the number one case used to argue that the business cycle could not be modelled as deviations from an exponential trend. By 1989 UK GDP was back to trend. Eyeballing annual data, I claim that the apparently supply side shock occured roughly on October 19 1987 -- black monday when stock markets crashed. The caused the lady who was not for turning to panic, turn and pump up the money supply.

    The case that the UK's problems were not simply insufficient demand was at least as strong in 1985 as it is now. The effect of the panic driven stimulus strongly challenged that case then.

    My guess is that the UK has been hit by the crisiss and then by Osborne -- two demand shocks do not make a supply shock, but together they distort estimates of potential GDP.

    1. Jon, I actually think the oil price hikes in 2010-12 were a sizeable supply shock.

    2. «By 1989 UK GDP was back to trend»
      «the oil price hikes in 2010-12 were a sizeable supply shock»

      It seems fascinating to me that in this discussion about the long term trends of the UK economy, neither the blogger, nor the present two commenters dare to mention that in 1982 it became an oil exporter from the Scottish offshore oilfields and in 2007 it has become an oil importer:

      Both the inception and end of the oil boom have been suppy shocks, the first a positive one, the second a negative one.

      Or discuss the UK labour market without mentioning that there have been 2 million immigrants in 10 years.

      As to the impact of Scottish offshore oil on the UK's Tony Blair, not an idiot, wrote in 1987 (yes, 10 years before becoming a tory Prime Minister, wrote:

      «Mrs Thatcher has enjoyed two advantages over any other post-war premier. First, her arrival in Downing Street coincided with North Sea oil. The importance of this windfall to the Government’s political survival is incalculable. It has brought almost 70 billion pounds into the Treasury coffers since 1979, which is roughly equivalent to sevenpence on the standard rate of income tax for every year of Tory government. Without oil and asset sales, which themselves have totalled over £30 billion, Britain under the Tories could not have enjoyed tax cuts, nor could the Government have funded its commitments on public spending. More critical has been the balance-of-payments effect of oil. The economy has been growing under the impetus of a consumer boom that would have made Lord Barber blush. Bank lending has been growing at an annual rate of around 20 per cent (excluding borrowing to fund house purchases); credit-card debt has been increasing at a phenomenal rate; and these have combined to bring a retail-sales boom – which shows up dramatically in an increase in imported consumer goods. Previously such a boom and growth in imports would have produced a balance-of-payments deficit, a plunging currency and an immediate reining-back on spending, with lower rates of growth.

      Instead, oil has earned foreign exchange and also produces remittance payments from overseas investments bought with oil money. The situation is neither stable nor healthy in the long term: but in the short term it allows the living standards of the majority to rise rapidly, even though the industrial base, the ultimate foundation of a successful economy, is still only achieving the levels of output of 1979. The fact that we have failed to use oil to build a productive and modern industry for the future is something historians will deplore. Nevertheless, oil has been utterly essential to Mrs Thatcher’s electoral success. Academics and commentators may ruminate on the Thatcher ethos and its effect on social attitudes, but the voters are looking in their pockets»

      Forget that, and the last 30 years of UK economist history become a fairy tale.

    3. To add to my «Forget that, and the last 30 years of UK economist history become a fairy tale» I have just found this amazing celebration of Thatcherism:

      where the political and economic impact of Scottish offshore oil exports is not mentioned *at all*. The more things change the more they stay the same :-).

  5. Simon,

    Re the idea that output was “unsustainably high” prior to the crisis, I have doubts for the following reasons.

    If you believe in the Phillips curve relationship between inflation and unemployment, then output can be unsustainably high in the sense that where demand is excessive, the penalty will be excess inflation. But there was no excess inflation prior to the crisis.

    Alternatively if you believe in the NAIRU relationship between inflation and unemployment, then “unsustainably high” output is near impossible. Reason is that according to the NAIRU theory, inflation ACCELERATES, when demand is excessive.

    Re the output gap, I share Jon Livesey’s puzzlement (see above).

    1. @ Ralph Musgrave

      "If you believe in the Phillips curve relationship between inflation and unemployment, then output can be unsustainably high in the sense that where demand is excessive, the penalty will be excess inflation. But there was no excess inflation prior to the crisis."

      No consumer price inflation, but other kinds of inflation there were, I wonder if there is a kind of "communicating vessels" principle at work...

  6. I can't help wondering whether a lack of long term vision within companies is at least partially to blame. It seems to be particularly hard in the UK to get investment for anything that doesn't have a product result in the next year or so. This has been going on for years, but if anything is getting worse. It's hard to get productivity improvements when investment isn't occurring. I've no idea why this is happening, or why the share price increases for firms that cut investment. Long term improvements are difficult with short term vision.

  7. When I speak w/ US bankers, they constantly complain that regulators are preventing them from lending more. When I ask how regulators are impeding them, I get vague answers. I'm sure that regulators are generally more critical of bankers these days, probably with good reason, but also without much strategy.

    But when I see investments pitched to banks, the answer is usually 'no'. I sense the real declination reason is mostly fear. A generation of bankers can't forget the credit collapse in 2009. We don't have a statistical measure of phobia.

    The other problem is that very low inflation impedes investment. Have you ever tried to pitch an investment that isn't risk free with a 5% return both real and nominal? It won't be approved and it will harm your career. Now if it were a 10% nominal return and 5% real return, it will probably be approved. There is a psychological barrier to investing in single digit nominal returns. Furthermore, inflation effectively creates a buffer to absorb unexpected credit losses. I know this is contrary to financial textbooks where investors always think in real terms, but it is the way the world works.

  8. "but there was no indication of any overheating in the labour market either."

    This seems a surprising generalisation. There probably was on a sectoral basis. Low skilled labour has been in low demand because of technological change and high migration rates of low skilled low wage labour. Employers, however, say they cannot get skilled labour (which often comes outside the EU and now subject to strict quotas). Of course some of the biggest and most profile excesses have been banking. This sort of statement requires a careful analysis of the labour market as the macro expansion being pushed here could just aggravate the distortions.

    1. «Employers, however, say they cannot get skilled labour»

      Employes *always* say that they cannot killed labour, or labour at all. They would be stupid to say otherwise, and their communications are entirely focused on their self-interest.

      And they are technically right if the labour they can get needs to be paid more than subsistence, because that's labour marginal cost :-).

  9. if we could produce all that stuff then we werent living beyond our means (except insofar as we are depleting or destroying natural resources)

    so therefore the only logical answer is that the problem was we were lending to people instead of paying them

    increase peoples income and there would have been no crash