Thursday, 10 April 2014

The Central Banker of All Central Bankers Explains The Way To Recovery

"Moreover, why should we trust the analytical paradigms and macroeconomic models that led us so badly astray pre-crisis - those very models of which the shortfall of demand view is one offspring?" 

The transcript of a speech (including the quote above) held yesterday by Jaime Caruana, the head of the Bank for International Settlements, has just been published. Coming from a central banker, his views and conclusions are surprising to say the least. Why? Because many of his views run contrary to those held by most central bankers around the world and are instead much in line with, and based on, those promoted by Austrian economists and the Austrian Business Cycle Theory (ABCT)

Before you get too carried away, Mr Caruana still advocates management of the business cycle by central banks and the bailing out of banks in the midst of financial crisis. Where his views differ from those normally communicated by central bankers are his interpretations of the crisis and policy prescriptions. 

In the paper, he discusses two main views on the causes of "chronic [economic] weakness" experienced in many countries following the 2008 bust and what should be done about it going forward. 

He largely, though not completely, dismisses the first view which he refers to as the "shortfall of demand view". In his own words, this view "argues that there is a persistent global shortfall of demand, which is particularly acute in crisis-hit advanced economies". This is the view held by most (all?) central bankers and mainstream economists today in my experience. 

The second view he puts forward is the view that "maintains that the key is removing the impediments to growth that are the legacy of a balance sheet recession". He refers to this view as the "balance sheet view". 

Before explaining these two views in detail, he describes three "takeaways" from the article (my bold):
  • First, not all recessions are born equal. The post-crisis recession we have seen is a balance sheet recession, linked to the bust phase of a long financial cycle. The unsustainable pre-crisis financial boom created a false sense of prosperity while masking balance sheet vulnerabilities and real resource misallocations, both across sectors and in the aggregate. As the boom turned to bust, the subsequent balance sheet recession revealed and exacerbated these weaknesses. In addressing a balance sheet recession, the key is to target the nexus between debt overhangs, poor asset quality and the misallocation of both capital and labour. These problems cannot be redressed by traditional aggregate demand policies alone. If not sufficiently redressed, the balance of risks and rewards of aggregate demand policies – monetary policy in particular – will worsen over time. Strong sustainable growth requires tackling these impediments head-on and restoring productivity growth.
  • Second, the post-crisis policy response has been unbalanced. There has been a lot of emphasis on fiscal pump-priming and monetary accommodation and not enough on balance sheet repair and structural reforms. This suggests that we have not yet fully come to grips with the financial cycle. In economies still recovering from balance sheet recessions, reacting ever more aggressively with monetary and fiscal policy will not resolve the problem. After a certain point, it may even be counterproductive (eg depletion of policy ammunition, development of new imbalances).
  • And third, given the previous takeaways, a logical policy conclusion would be that policy frameworks need to adjust based on a longer-term focus. Monetary, fiscal and prudential policies alike need to be more symmetrical over the financial cycle: policies should lean more deliberately against financial booms and ease less aggressively and persistently during busts, assigning priority to repair and reform. This symmetry would limit the amplitude of the disruptive cycles and avoid the risk of exhausting ammunition over time. Correspondingly, there should be more emphasis on structural policies (eg labour, competition, taxation) to address structural problems. There are no short cuts: the path from here to there will not be easy. But the longer we wait, the bigger the costs will be.
These three points alone should, at least partially, please readers frustrated with governments and central bankers focused on policies that "stimulate aggregate demand". 

But there's more of the good stuff to come. Caruana explains "I shall argue that the balance sheet view provides a more convincing overarching interpretation of what has been happening in the global economy in general..." and
But, equally, there is no doubt in my mind that, as an overarching explanation, the balance sheet view is much more compelling. It can explain in one single sweep both pre- and post-crisis developments across the world. It served us well pre-crisis in identifying risks as events unfolded. And it seems to trace them better now, for both crisis-hit and non-crisis-hit economies. Surely Occam's razor applies. Moreover, why should we trust the analytical paradigms and macroeconomic models that led us so badly astray pre-crisis - those very models of which the shortfall of demand view is one offspring?
Therefore, instead of focusing on his refutations and arguments against the first view (the shortfall of demand view), I will here lay out his arguments in favour of the second view (the balance sheet view). 

Caruana summarises the balance sheet view in seven propositions (my bold): 
  1. First, the crisis is not an exogenous shock; rather, it represents the inevitable collapse of a previous unsustainable boom – the bust phase of a financial cycle. This naturally ushers in a balance sheet recession, featuring large sectoral and aggregate debt and capital stock overhangs and an impaired financial sector.
  2. Second, given initial conditions, deleveraging is a necessary precondition for a self-sustained recovery: debt overhangs relative to income need to be reabsorbed for the economy to rebound sustainably.
  3. Third, a key factor restricting aggregate demand is anaemic credit demand: agents realise that they have borrowed too much given their lower income expectations and seek to pay down debt. This numbs the effectiveness of monetary and fiscal stimulus.
  4. Fourth, the major asymmetry in the global monetary and financial system that worsens the picture is that easy monetary conditions in major economies spread to the rest of the world. As a result, the system has produced inappropriately low interest rates for the world as a whole. Non-crisis-hit countries find it hard to operate with interest rates that are significantly higher than those in the large crisis-hit jurisdictions because of the fear of exchange rate overshooting, even when the economy has been growing strongly. In several cases, this has been fuelling unsustainable financial booms (“financial imbalances”). The result is expansionary in the short run but contractionary over the longer term.
  5. Fifth, a key source of hysteresis is the misallocations of credit and resources (capital and labour) that built up during the unsustainable boom and may worsen during the bust. Given debt overhangs, in crisis-hit economies the allocation of credit matters more than the total amount of credit extension for the recovery and medium-term growth.
  6. Sixth, negative real interest rates, especially when associated with zero policy rates, are not equilibrium phenomena. As a result, they risk causing collateral damage, not only in the crisis-hit countries themselves, where they may further delay balance sheet adjustment or encourage unhealthy forms of risk-taking, but also, and more visibly, elsewhere in the world, by causing a build-up of financial imbalances. This, in turn, could end up validating those low interest rates, as the unwinding of the imbalances could make normalisation extraordinarily difficult globally.
  7. Finally, disinflationary, and possibly deflationary, pressures may in part reflect benign underlying forces, notably heightened global competition, that the boom had obscured. To that extent, the concern is not so much declines in the prices of goods and services per se as the harmful interaction between debt and asset prices.
Students of Austrian economics and the ABCT will recognise and agree with many of Caruana's propositions above as he is clearly laying out that the bust was due to an unsustainable boom. He also explains that a recession is necessary to correct these mistakes ("deleveraging"). In addition, he also recognises that artificially low interest rates may encourage "unhealthy risk-taking", i.e. the creation of new bubbles. Finally, Caruana regognises that not all deflation is inherently bad (he also later in the article writes "The historical evidence indicates that deflations have often been associated with sustained growth in output".)

He proceeds by discussing the evidence supporting the "balance sheet view" (my bold),
In contrast [to the "shortfall of demand view"], the balance sheet view is more controversial and not part of the prevailing paradigm. To my knowledge, there is not yet a macroeconomic model that captures this view. Nonetheless, there is considerable evidence that supports it – including evidence based on research by my colleagues at the BIS. Let me outline here some evidence in six areas.
First, the nexus between systemic banking crises, financial cycles and the measurement of potential output. Recent studies have found that banking crises, and the balance sheet recessions they usher in, tend to occur at the peak of financial cycles. Such cycles are best captured by the joint behaviour of credit and property prices and, crucially, are much longer than business cycles: they last from 16 to 20 years rather than up to eight or 10 years. Indeed, it is possible to construct fairly reliable real-time leading indicators of systemic banking crises based precisely on the behaviour of credit and property prices. These also identified fairly accurately the vulnerabilities ahead of the recent crisis.
Second, the measurement of potential output. Systemic banking crises have particularly persistent effects on output and, in all probability, potential output too. It is now widely accepted that they have historically been associated with large permanent output losses: output may at some point grow again at its pre-crisis long-term rate, but does not return to its previous path. The current episode is no exception. Unless one is willing to entertain the (implausible) idea that output can be below potential for decades, this indicates that potential output is lower than estimated pre-crisis.
Additional evidence supports this point; and it also casts doubt on more traditional measures of potential that generally draw on the behaviour of inflation. Importantly, all the measures of potential output used nowadays in policy institutions and academia failed to identify in real time that output was running above its potential, or sustainable, level ahead of the great financial crisis: after all, inflation was low and stable. They have done so only ex post, a few years after the recession set in, as those institutions and scholars rewrite history with the benefit of hindsight. By contrast, recent work indicates that, by using information from the financial cycle, notably the behaviour of credit and property prices, it is possible to do precisely that. The intuition is simple: it was the build-up of financial imbalances, not rising inflation, that signalled the unsustainable expansion of output before the crisis. Such financial booms may also have helped to mask the trend decline in productivity growth, while globalised competition helped to keep goods prices in check.
Third, the role of deleveraging. There is evidence that, in sharp contrast to a normal recession, a balance sheet recession requires deleveraging. In this case, deleveraging during the recession induces a stronger subsequent recovery. Moreover, there is generally little link between post-crisis credit expansion, beyond the recession, and output growth: so-called “credit-less recoveries” are the rule, not the exception, following balance sheet recessions. This casts the post-crisis slow bank credit growth in a different light. In addition, in contrast to the view that stresses regulation-induced credit supply constraints, much of the adjustment in the banking sector has taken place through retained earnings and external capital, and far less through lending cuts. Likewise, a large widening of lending spreads has been persistent mainly in some peripheral euro area countries.
Not surprisingly, for much the same reason, balance sheet recessions, in contrast to normal ones, appear to numb the effectiveness of both monetary and fiscal policy. A recent cross-country study suggests that looser monetary and fiscal policies during the recession are not followed by stronger subsequent recoveries. This is consistent with the intuition that agents’ attempt to pay back debt reduces the impact of lower interest rates and fiscal transfers: agents have little intention to borrow more and prefer to save the additional income. The result is not surprising for monetary policy, but conflicts with empirical studies that suggest fiscal multipliers are higher when economic slack is large. A possible explanation is that those studies fail to distinguish normal from balance sheet recessions.
Fourth, the source of hysteresis. The risk of hysteresis as a result of credit and real resource misallocations is borne out, in particular, by the Japanese experience. There is considerable evidence that, following the financial bust in the early 1990s, the combination of very low interest rates and forbearance hurt production potential, by trapping resources in inefficient companies at the expense of more profitable ones. Nordic countries, which faced similar problems roughly at the same time, and were forced to take tougher monetary and fiscal policies, recovered much better. Presumably, in addition to larger exchange rate depreciations, this reflected in part their aggressive steps to repair the banking system and reform the economy. Forbearance and delays in balance sheet repair also appear to have been at play in the more recent post-crisis experience, at least in some jurisdictions.
Fifth, the causes of deflationary pressures. The historical evidence indicates that deflations have often been associated with sustained growth in output. Surprisingly perhaps, the Great Depression was more the exception than the rule. Similar spells have also been evident since the 2000s, including in China, Nordic countries and, as I speak, Switzerland. In these cases, sustained output growth has proceeded alongside strong increases in credit and asset prices. At the same time, there is growing evidence that global factors have been playing a bigger role in domestic inflation just as domestic measures of slack have lost significance. Indeed, macroeconomists and policymakers know that, for many years now, the link between domestic slack and inflation has proved elusive. To be sure, one common explanation is that better-anchored expectations, underpinned by greater central bank credibility, have reduced the sensitivity of inflation. Even so, it would be implausible to rule out the role of the globalisation of the real economy, notably the entry of China and former communist countries into the world trading system. Surely, in addition to relative prices changes, this produced welcome disinflationary tailwinds, which were at play well before the crisis and may well still be at play.
Finally, the asymmetries in the international monetary system. There is considerable evidence that, for the world as a whole, policy interest rates have been persistently below traditional benchmarks, fostering unbalanced expansions. Policy rates are comparatively low regardless of the benchmarks – be these trend growth rates or more refined ones that capture the influence of output and inflation, such as Taylor rules. Moreover, there is clear evidence that US monetary policy helps explain these deviations, especially for small open and emerging market economies. This, together with the large accumulation of foreign exchange reserves, is consistent with the view that these countries find it hard, economically or politically, to operate with rates that are considerably higher than those in core advanced economies. And, alongside such low rates, several of these economies, including some large ones, have been exhibiting signs of a build-up of financial imbalances worryingly reminiscent of that observed in the economies that were later hit by the crisis. Importantly, some of the financial imbalances have been building up in current account surplus countries, such as China, which can ill afford to use traditional policies to boost domestic demand further. This is by no means new: historically, some of the most disruptive financial booms have occurred in current account surplus countries. The United States in the 1920s and Japan in the 1980s immediately spring to mind.
Caruana then lays out his policy prescriptions based on the "balance sheet view",
  • The priority is not so much mechanically to fill an output gap through traditional demand management. Rather, it is to establish the basis for a self-sustained and prompt recovery through aggressive balance sheet repair, resolving the legacy of the crisis and limiting the risk of chronic weakness. In this phase, monetary policy should work together with prudential and fiscal policies to address the debt overhang–poor asset quality nexus head-on. It should also give way to structural policy, to reduce other impediments to growth. And in deciding the necessary degree of accommodation, monetary policymakers should consider carefully to what extent disinflationary pressures result from positive supply developments or from a domestic shortfall in demand.
  • Failing to carry out these policies raises several risks, as standard aggregate demand measures fail to gain lasting traction. Fiscal policy expansion risks undermining further the sustainability of public sector finances. A persistently aggressive monetary policy risks exacerbating collateral damage, both domestically and internationally, as unwelcome spillovers foster the build-up of disruptive financial imbalances in other countries whenever financial cycles are out of sync. And as results disappoint, such a policy can ultimately sap the central banks’ credibility, effectiveness and public support. 
  • More generally, there is a serious risk of exhausting the policy room for manoeuvre over time. As policymakers respond asymmetrically over successive business and financial cycles, hardly tightening or even easing during booms and easing aggressively and persistently during busts, they run out of ammunition and entrench instability. Failure to consider the sources of disinflationary pressures can add to this risk. As a result, lasting normalisation remains elusive. In particular, the accumulation of debt and the distortions in production and investment patterns associated with unusually low interest rates hinder their return to more normal levels. Low rates, paradoxically, validate themselves. 
  • This is an instance of “time inconsistency”: policy steps look compelling when taken in isolation, but, as a sequence, they lead policymakers astray. The shortfall of demand view sees no tension between boosting demand in the short term and longer-term growth; if anything, the former promotes the latter. The balance sheet view is fundamentally about intertemporal trade-offs: a short-term approach to boosting demand risks more wrenching pain further down the road. The only way to reconcile short-term and long-term prospects is to improve the latter, so as to unblock private demand, especially investment.
Furthermore, he explains that "the balance sheet view has implications for policy responses at the current juncture and for adjustments to policy frameworks",
  • Policies now should take advantage of the window of opportunity provided by stronger growth to repair and reform. Monetary policy should stay on course towards normalisation, avoiding the risk of financial dominance, ie of being unduly influenced by financial market jitters. By filling the void left by other policies, it has been overburdened for too long. Fiscal policy should keep a close eye on long-term sustainability: it should consolidate where fiscal trajectories are unsustainable and make room to use any available firepower to support the restructuring of balance sheets, such as through banking system recapitalisations. Moreover, it should not be lulled into a false sense of security where unsustainable financial booms have been under way: there is ample evidence that they flatter the fiscal accounts, as potential output growth is overestimated, revenues are bloated and contingent liabilities accumulate. Prudential policy should strengthen financial institutions’ balance sheets. In crisis-hit economies, this means full and prompt balance sheet repair, by enforcing loss recognition and building up capital and liquidity buffers. In those economies experiencing financial booms, it means strengthening defences further. Everywhere, it means completing the financial regulatory reforms. And it means paying much more attention to structural policies that raise growth potential. This would allow countries to rebound more strongly post-crisis and reverse the long-term trend decline in productivity growth.
  • Policy frameworks need to incorporate financial cycles systematically. As my colleagues and I have explained in more detail elsewhere, policies – monetary, fiscal and prudential – should respond more deliberately to financial booms, by building up buffers, and respond less aggressively and persistently to busts, by drawing the buffers down. This calls for longer policy horizons than those currently in place – recall that the financial cycle is much longer than the business cycle. And it calls for governance arrangements that effectively insulate policymakers from the huge political economy pressures that induce asymmetric policies: no one objects during a boom; everyone demands support during a bust. These adjustments, unfortunately, are easy to identify but exceedingly difficult to implement.
Finally, in his concluding remarks he notes,
The balance sheet view highlights the tight nexus between balance sheets and economic activity. It focuses on how balance sheet problems and the associated distortions in the real economy develop during financial booms, hardly noticed under the rising tide, only to emerge during busts, once the tide recedes. It acknowledges the power of financial factors, but recognises that only structural ones hold the key to long-term growth. It acknowledges that the interaction between financial factors and monetary policy influences inflation and disinflation, but it recognises that, in today’s highly integrated world economy, the globalisation dividend may still be generating welcome disinflationary pressures. It stresses stocks, not only flows. It emphasises policies with a financial cycle perspective, not individual responses to isolated shocks.
The policy prescriptions follow naturally. Where economies are still recovering from balance sheet recessions, address head-on the pernicious nexus between the debt overhang and poor asset quality and rely less on traditional aggregate demand policies: they are less effective and may possibly, at some point, become counterproductive in the longer term. After all, not all recessions are born equal. Where economies have been experiencing financial booms, lean aggressively against them and do not be fooled by their false veneer of prosperity. After all, not all expansions are equally sustainable. Since initial conditions matter, everywhere, make sure that fiscal accounts are sound and financial institutions well capitalised, and work hard on structural policies. After all, there are no short cuts. Finally, adjust policy frameworks so that monetary, prudential and fiscal policies are more symmetrical over successive financial and business cycles. This would reduce the probability and severity of endogenous financial crises and provide policymakers with more leeway to act.
In summary, Caruana explicitly, though not officially, recognises many of the lessons from Austrian Business Cycle Theory (ABCT) and also puts forth some solutions to end the chronic lack of growth which I think many adherents of Austrian economics would welcome. As such, these points of view coming from perhaps the central banker of all central bankers is good news. It will hopefully receive attention in mainstream news and help to create a healthy debate among mainstream economists, monetary cranks and demagogues. Perhaps Caruana's speech and paper also can serve as an excuse for central bankers to implement policies which are less damaging than those currently being recycled around the world.

Read his full speech titled "Global economic and financial challenges: a tale of two views" here.