Reserve Bank of Australia governor Glenn Stevens delivered his annual Anika Foundation speech at lunchtime today.
Stevens is on the Board of the foundation, which raises funds for research into adolescent depression and over the last few years, it become one of the RBA Governor’s most important thematic speeches. Here’s what he said on the challenges for economic policy:
The challenges I want to focus on are those involved in dealing with and then recovering from the financial crisis that enveloped the major European countries and the United States (US) (and, briefly, the whole world) in late 2008. My view is that policies were very effective in averting a potential catastrophe five years ago. But fostering a strong recovery has been much more difficult. I think this reflects the nature of the shock policymakers have been dealing with. We have learned, once again, about the limits to what monetary policy alone can achieve.
I wish to be clear at the outset that my remarks today are about global issues, and contain no particular message specific to Australia.
I begin with some remarks on the crisis itself.
Stabilising the Crisis
I’ve observed a number of economic and financial episodes over the past 30 plus years. In just about every one of them of any significance, people have claimed it was the worst episode since the 1930s. One of the earliest substantive pieces of research I was involved in at the Reserve Bank sought to bring some evidence to bear on the claim that the recession in 1983 was somehow comparable with the 1930s. It wasn’t, of course – not even close. Almost all such claims are greatly exaggerated.
But when those who lived through 2008 and 2009 say that there was the potential for an outcome every bit as disastrous as the 1930s, I don’t think that is an exaggeration.
Any account of the events of September and October 2008 reminds one of what an extraordinary couple of months they were. Virtually every day would bring news of major financial institutions in distress, markets gyrating wildly or closing altogether, rapid international spillovers, and public interventions on an unprecedented scale in an attempt to stabilise the situation.
It was a global panic. The accounts of some of the key decision-makers that have been published give even more sense of how desperately close to the edge they thought the system came, and how difficult the task was of stopping it going over.
One ‘advantage’ the current generation of policymakers had was that, although they had never lived through a 1930s-style episode, they had at least some understanding of what had happened, and had absorbed the lessons of the sorts of things to do, and to avoid, once a crisis erupted. This was a result of a large output of scholarship, including by one scholar who happened to end up being Chairman of the Federal Reserve Board at the critical moment.
Among those lessons, though this is hardly an exhaustive list, are the following:
Policymakers understood these lessons. There were naturally mistakes and misjudgements made along the way – that will always be the case. But by and large, these lessons informed the response to the crisis.
Liquidity in most jurisdictions was ample. Central banks were, when needed, quite generous and inventive supplying it. A point made many times was that liquidity cannot redress a solvency problem. That’s true, but not providing liquidity to a stressed system will ultimately make solvency problems worse.
Authorities in all jurisdictions understood the need to prevent a spiral of defaults of large financial institutions further adding to already tightened credit conditions. In October 2008 the G7 finance ministers and central bank governors said pretty explicitly there would be no more failures of systemically important entities. Countries were prepared to use guarantees, and there was public injection of capital into banks in a number of instances.
People will rightly worry about moral hazard and perpetuating ‘too big to fail’ and so on. That’s a legitimate concern and a powerful reason for reform for the future, but it wasn’t a basis for policy when facing imminent disaster.
As for macroeconomic policy, just about every country adopted more expansionary settings. The major central banks, and most smaller ones, reduced interest rates aggressively after the 2008 failure of Lehman Brothers. Governments were advised by international bodies like the IMF to pursue fiscal expansion, to the extent they had ‘fiscal space’. Most of them did, even in cases where ‘space’ was rather limited.
The spirit of international cooperation, so often honoured in the breach in normal times was, at critical moments, actually pretty good. For example, the swap lines between central banks were put into place quickly, and around US$500 billion in US dollar liquidity flowed into the global system during the last quarter of 2008.
Agreement on the broad contours of other elements of the responses at the height of the crisis was found fairly quickly. At moments when the world needed to hear that the people in charge had a plan, the G20 Leaders sounded like they had one. As well as stabilising the financial system, they committed to avoiding protectionism and eschewed competitive exchange rate devaluations.
As the Leaders said at the September 2009 Summit in Pittsburgh: ‘it worked’. The decline in global output and trade in the last quarter of 2008 was at a pace that rivalled the contraction in the 1930s. Had it gone on, we can be sure that tens of millions more people would be unemployed and trillions of dollars more wealth would have been destroyed.
But it didn’t go on. It was arrested. As a result, we talk about the ‘Great Recession’, but we don’t talk about the ‘Great Depression Mark II’. These initial interventions achieved what they were supposed to. We might not like the politics or the optics of it all – all the ‘bailouts’, the sense that some people who behaved irresponsibly got away with it, the recriminations, the second-guessing after the event and so on. But the alternative was worse. I think all that would be more or less accepted by most reasonable people.
But what about the efficacy of policies in generating a robust recovery? It is that set of issues to which I now want to turn.
Fixing the Banks
It is one thing to staunch the tide of bank failures by public intervention. But to help secure a recovery in the financial system and the economy, banks’ balance sheets have to be fixed. There were genuine problems with asset quality and it would have done little good to pretend otherwise. There had to be an honest accounting of these problems. Until there was, counterparty risk aversion would persist, because no-one knew whom to trust.
An honest accounting of the assets meant owning up to the shortage of capital. There never had been enough capital for many entities to have run the risks they carried, and what capital there had been was in many instances reduced or gone. So that had to be fixed, either by raising capital from the markets or accepting it from the government.
The stress tests conducted by the US authorities in 2009 were designed with this dual purpose in mind. They were effective in getting clarity on asset quality and in strengthening capital positions. This has been harder to achieve in Europe, for various reasons to do with the complexity of the European project, and remains something of a work in progress. Nonetheless, the lesson would still hold that fixing the balance sheets is a necessary condition for recovery.
But fixing the banks, while a necessary condition, isn’t a sufficient one. A sound financial system can accommodate growth by supplying the necessary credit, but a sound financial system alone isn’t the initiating force for growth – at least, not the kind of soundly based growth needed. Macroeconomic policies have a role too.
One of the difficulties has been that public debt burdens rose sharply. This was partly as a result of the cost of fiscal stimulus measures and bank recapitalisations in some cases, but it was mainly because of the depth of the downturn in economic activity. A financial crisis and deep recession can easily add 20 or 30 percentage points to the ratio of debt to GDP, and did so in a number of cases.
Moreover, since it appears that economic activity isn’t going to be back on its previous trend levels any time soon in many of the crisis-affected countries, and hence government revenues seem to be persistently on a lower track, public finances have been left in a fragile condition.
That some countries went into this episode having avoided serious efforts at fiscal consolidation for quite a long time, and hence already had significant debt burdens, only heightened the problems.
So fiscal policy has not had as much scope to continue supporting recovery as might have been hoped. Policymakers in some instances have felt they had little choice but to move into consolidation mode early in the recovery.
That has left monetary policy trying to support demand, but monetary policy has had its own difficulties in doing so. The Fed and some other central banks quickly found themselves in a situation in which the short-term interest rate reached, for practical purposes, the ‘zero lower bound’. This has echoes of the ‘liquidity trap’, something we had learned about as a textbook possibility, but which now seemed to have become a real-life experience. This had already been noticed for Japan in a memorable paper by Paul Krugman as long ago as 1998.
The liquidity trap was a hypothesis of Keynes, who contemplated the possibility of a world in which additional injections of money into the economy become ineffective at reducing interest rates, because people become content simply to hoard the extra cash. In the original specification this was supposed to happen at some positive interest rate. In fact, the short rate could fall all the way to zero. But it couldn’t realistically get much lower.
That still left long rates, though, and in countries like the US and the euro area these rates matter a great deal. The central banks did not accept that the policy rate reaching zero meant the end of effective easing in policy. Enter ‘quantitative easing’, or ‘QE’.
QE is essentially unsterilised intervention in asset markets, meaning that the central bank expands its own balance sheet by buying assets for cash. The market for long‑term government debt, agency securities and some types of private paper all were affected by QE or similar measures at one time or another in the United Kingdom (UK), the US and of course Japan. Such interventions also occurred in Europe, though with slight differences in the motivating narrative, given the particular circumstances of the single currency arrangements.
By definition, since it is unsterilised, QE results in a rise in settlement balances at the central bank. Sometimes people point to this as indicating a failure of the policy, as though banks have ‘failed to lend out the extra cash’. But this isn’t an accurate description of what has happened. Those extra settlement balances have to remain in the system, however hard the market participants try to lend them out to others, until the central bank decides to take them out again.
In fact, QE mainly works not by some credit multiplier mechanism, but by pushing down the cost of capital for the economy. Long-term rates fall below where they would otherwise be as the central bank bids for securities in the market. The former holders of those securities then have to re-deploy the cash they now hold. Unless they are all content simply to absorb extra cash – unless, that is, the liquidity trap is fully in operation for all asset classes at all maturities – their efforts to acquire other assets with similar duration but somewhat higher risk will bring down yields on those other assets (including foreign assets), lowering risk premia and so on. This ‘search for yield’ means that financial conditions, broadly defined, ease: that is, the cost of capital to the non-financial sector declines.
It has been noted that this amounts to more risk-taking among investors. Indeed it does. While in some discussion it seems to be implied that this is a bad thing, actually prompting more risk-taking is the whole point. Assuming we accept the notion that there is a role for stabilisation policy, when appetite for risk evaporates its job is to respond in a way that helps to restore that appetite, up to a point.
The key question is: what kind of risk-taking? In particular, to what extent is the lower cost of capital to the non-financial sector resulting in more risk-taking in that sector? To what extent are businesses in the ‘real economy’ becoming, as a result of accommodative monetary policies, more prepared to take a risk – on a new product, a new factory or process, an innovation, a new market or a new employee?
For if some increased risk in the financial sector is part of the process of getting more genuine entrepreneurs in the economy to take the sorts of risk that are part and parcel of restoring the dynamic of growth, that is probably a trade-off worth making. On the other hand, if some years from now we find ourselves looking back and concluding that such ‘real economy’ entrepreneurial risk-taking has not really taken place, and all that has happened is that financial risk-taking and leverage have risen, we would be disappointed.
Unfortunately, it is very difficult, at this stage, to evaluate how well the trade-off is operating. It is certainly clear that, globally, financial conditions have been extraordinarily accommodative and that the ‘search for yield’ has been a pervasive force. Risk spreads are low, overall costs of borrowing are exceptionally low, and yields on all manner of assets are being bid down to very unusual levels, around the world. But as for the effects filtering through to risk-taking in the real economy, I think the only honest assessment is that the evidence is hard to read. The US economy, where the QE actions have been the most aggressive, certainly has seen recovery, though at only a moderate pace. Business capital spending, perhaps a high level gauge of ‘real economy’ risk-taking, has risen in the US, but has hardly been strong and it remains quite weak in several other major economies.
Some would take that to indicate that the unconventional monetary policy has not been all that effective, and too risky; others would see it as a sign that policy did not try hard enough. Still others, myself among them, remembering that it always takes time for an economy to heal after a financial crisis, and that as usual we don’t know the counterfactual, might simply feel that it is impossible to draw strong conclusions.
Limits to Monetary Policy?
One cannot help but observe, though, that monetary policy’s ability to help growth can be impaired by things that the central bank cannot influence. One of monetary policy’s key effects is thought to come by lowering the cost of borrowing relative to the expected rate of return on real capital – the so-called Wicksellian ‘natural rate’. But if for some reason the natural rate is very low, then for monetary policy to impart this expansionary impetus to the economy requires very low policy rates. Perhaps the natural rate could for some reason fall so low that it becomes impossible for monetary policy to impart very much stimulus, because a wide range of nominal interest rates approach or reach the zero bound and that is not low enough. In extremis, a persistent case of this would amount to conditions that might be described as ‘secular stagnation’, as recently discussed by Larry Summers.
In that sort of world, central banks can still carry out some of their critical functions. They can make portfolios more liquid, and cut short an incipient spiral of falling asset values, contracting credit and so on that might otherwise be much more devastating. They can significantly lessen the burden on those with too much debt as they seek to deleverage, admittedly at the cost of lowering the incomes of savers. But if people simply don’t wish to take on new business risks, monetary policy can’t make them.
What sorts of things might result in a very low ‘natural’ rate of return on real capital? And have they been in operation?
The Austrian tradition would point to previous over-investment, after which, expected returns are perceived to be low. That seems to have been part of the story of Japan after the ‘bubble economy’ era. In the more recent episode, there was clearly some over-investment in the housing stock in the US and one or two other countries (Spain, for example). But a casual look at business investment in most advanced countries does not suggest obvious over-investment prior to 2008. Visitors to the US would not readily conclude that America had over-invested in infrastructure (nor would they when visiting the UK or, for that matter, Australia).
A decline in an economy’s potential growth rate, driven by lower population or productivity growth could be a factor. Japan has a declining population and the dynamics of some of the European countries are not much better. But this isn’t obviously the case for a country like the US.
If the process of innovation that is a precursor to productivity growth is impaired, or if research and experimentation simply has periods of fewer discoveries, just by chance, that could also lessen the perceived opportunity to deploy capital profitably. But since 2008, the number of patents granted in the US has risen by nearly two-thirds. While only a crude measure, that doesn’t suggest the desire to innovate has collapsed.
Perhaps the answer is simply subdued ‘animal sprits’ – low levels of confidence. After all, the natural rate is an expected rate. If people think, for whatever reason, that returns for future possible investments will be low, or subject to high risk, then they will be reluctant to invest even if past and current returns are quite satisfactory. Conceivably, this could be a self-reinforcing equilibrium.
Assessing this is difficult and there is the potential for multiple equilibria. But it does seem, to me, that businesses in many places are much more conscious of risk, relative to return, than they were in the pre-crisis period. At some stage, hopefully share market analysts and the investor community will ask fewer questions about risk reduction, and more about the company’s growth strategy.
Suppose some such forces have, in fact, been in operation. Will we be stuck in a low-return, pessimistic equilibrium for a long time? Or will such forces abate, and if so, when?
I would argue for realism, as opposed to either naïve optimism or determined pessimism. Certainly earlier expectations about risk and return were too optimistic. We have been through a period of adjustment. But it’s doubtful that the desire to experiment and innovate has entirely disappeared. And it seems unduly pessimistic to think that everything that can be invented has been, or that every improvement to existing ways of doing things has already been implemented. And unless we think the tendency for human optimism has been completely drummed out of us, animal spirits in the ‘real economy’ will surely improve at some point.
The question is what might be done to help that happen more quickly. It is highly unlikely that the answer will come, in any country, from monetary policy. But this is where the G20 agenda on growth can, if it is well used, be of considerable help:
then the growth potential of the world will look, and actually be, improved.
The highly accommodative financial conditions will then have a more powerful effect in engendering real growth. A rising confidence dynamic could unfold. The prospects for profitable investments by businesses would be significantly improved.
And then, at a future event of this kind we would be able to conclude that the challenges facing economic policy had been met.
For now, it remains for me to thank you once again for coming out today to support the Anika Foundation’s work in raising awareness about adolescent depression and suicide.