Speech Observations on the Current Situation
Address to Credit Suisse Global Markets Macro Conference
New York –
It is good to be in New York once again. Thank you to Credit Suisse for the invitation.
By way of a preface to these remarks, I note that we have all just been in Washington, D.C. for the IMF and G20 meetings. With that context, my comments today are international in focus, rather than carrying any particular message about my own country.
It has certainly been another interesting few months in global financial markets. Faced with falling commodity prices, diverging policy stances among major jurisdictions, the odd policy surprise and the sense that the economic growth outlook was a bit softer than had been assumed before, market participants had a lot to digest. To this picture was added a new dimension of uncertainty about the way regulations pertaining to some capital instruments might affect returns, and about the way resolution actions may work in some places.
In the opening weeks of this year we saw sovereign yields in major advanced economies move towards historic lows, while spreads for some other sovereigns and corporates – especially high-yield obligations – moved out, with particularly sharp increases for bonds issued by resources companies. Share market volatility increased, with prices generally lower, especially for banks. Currencies of a range of emerging market countries, already marked down considerably over the preceding year, came under more pressure and sovereign spreads for some, though not all, of these countries increased.
Since about the middle of February, markets seem to have regained some composure. A range of commodity prices have risen. Share prices have recovered some ground, though they remain mostly lower than they were a year or six months ago, and bank share prices in Japan and Europe are back around their recent lows. Investment-grade spreads have declined to where they were about six months ago. That is higher, to be sure, than a year or two years ago, but those spreads were unusually compressed. Even energy-related spreads have narrowed quite noticeably, to be back around their levels at the start of the year. The same is true for emerging market sovereign spreads, while yields on advanced-economy bonds remain close to their record lows. Emerging market currencies have generally appreciated, as have the currencies of commodity-exporting nations.
So things have calmed down somewhat. That said, I think most observers and policymakers tend to the view that while some recovery is welcome, the relative ‘calm’ seems a little eerie – perhaps fragile. Certainly, these events have posed a few questions for policymakers, at least for the present speaker. Among the questions being considered over recent weeks were the following:
- Does the ‘turbulence’ simply reflect the same information that is embodied in the softer global growth forecasts that have been emerging over this period? In other words is it just a noisier version of a signal that is already being received?
- Or has the financial volatility been telling us there has been a significant shock, the effects of which are coming but which we can't see (yet) in forecasts or other data?
- Alternatively, could the turbulence be a shock that leads to a worse global outcome – by leading to a tightening of credit conditions, loss of wealth and confidence, etc., and therefore crimping demand?
- Or is it just a bout of market nervousness that carries little lasting importance?
As always, it is impossible to be sure, but it would not be unreasonable, I think, to draw the tentative conclusion that while these movements did reflect some underlying softening in the global outlook that was already emerging, the reaction was overdone. Commodity prices are well down, but actually some prices had been declining for quite a while. Iron ore, for example, peaked as long ago as February 2011. Moreover, in many cases declining commodity prices reflect additional supply, which usually carries a different – positive – implication for global growth, as opposed to weaker demand.
While global growth forecasts are being lowered, at this stage they see higher growth rates than in 2001, which was a relatively mild slowdown episode. Of course that is just a forecast – but so far there are not any actual data that invalidate that view. Given some recovery in markets of late, it seems too extreme to conclude that this event itself is developing into a significant financial shock with important additional macroeconomic significance beyond the softer global growth already understood to be in prospect.
Having said all that, I don't think we should write the episode off as just another bout of nervousness. Even if the volatility was not necessarily a reflection of fundamentals, it's worth ensuring that we have extracted all the information we can from it. In addition, some quite big policy questions are lurking. It may be that these are playing a role in bouts of market nervousness – and could do so again.
With that in mind, let me talk about a few issues – some familiar, but one or two rather newer.
One issue that both the regulatory community and many market participants have focused on for a while now is an apparent decline in market liquidity at a time when capital markets have become a more important source of funding for the economy. We can debate how much of that decline in liquidity is a result of regulation – and certainly some of it is – and how much simply owes to large players deciding to alter the way they run their businesses independently of regulatory changes.
There is at least some evidence that market liquidity is more costly than it used to be. Whether that is a big a problem is not clear. It could matter, for example, through the cost of capital to the productive side of the economy, as investors price the illiquidity of the assets they buy. This would mean that borrowers pay a little more than otherwise as a result. But are creditworthy borrowers actually having material difficulties accessing finance on reasonable terms, beyond what is explainable by their own macroeconomic fundamentals?
It also seems that liquidity is somewhat less reliable even under conditions that could be thought to be ‘normal’, even in some of the largest sovereign markets in the world. Recent events have not obviously demonstrated any deterioration, nor any improvement, in that situation.
We don't know how liquidity will stand up under genuine stressed conditions. This is of increasing importance, particularly given the large growth of assets under management. Regulators are rightly emphasising the need for sound liquidity management by the asset managers, while market participants rightly point to the practices and tools they have to address these questions. But the key question is: what liquidity promise do the investors – many of them retail – think they have been given? Not what the fine-print says, but what are they actually assuming? If their assumptions are at variance with the genuine underlying liquidity in the relevant markets, there could still be trouble even with careful planning by asset managers. The Financial Stability Board continues to work on these issues and over time presumably we will be able to refine our assessments.
Turning to other policies, when the financial crisis ushered in a very serious economic downturn in late 2008, central banks around the world reduced their policy rates dramatically – in a number of cases reaching the so called ‘zero bound’. Central bank balance sheet expansions were also enacted. These were initially associated with system-wide liquidity provision, but their more persistent use was an effort to provide more monetary policy support to weak economies. Together with more explicit guidance about the future path of short rates, these actions were aimed at reducing rates further along the yield curve. Some details differed, such as whether measures were configured to work through general capital market pricing or through the banking system, or indeed, through exchange rates. These details were determined by the institutional details of the various countries. But overall, ‘non-conventional’ policies had the effect of lowering borrowing costs by pushing savers further out on the yield and risk curves.
That is, they prompted a search for yield. They were supposed to do so. Portfolio substitution is part of how monetary policy works. When the central bank removes low-risk (and perhaps some not so low-risk) financial assets from the system as a monetary policy action, its intent is that, much further out along a long chain of substitution effects, consumers and businesses will respond by purchasing real goods and services.
The extent to which demand for real goods and services was increased as a result of these actions is hotly debated. In the United States, business investment as a share of GDP had largely recovered by 2015. Of course, that might have happened anyway. Meanwhile, investment is still below pre-crisis levels in Europe. In Japan, it has never recovered to levels prior to the collapse of the bubble in the early 1990s. Of course, those levels may well have been unsustainably high, and we can't know the counterfactual. So the debate continues.
My personal view is that central bank policy at the global level was very effective at heading off a potential catastrophe after the Lehman event, but was always going to have limited capacity to accelerate the recovery. That is not to say central bank policy should not have done all it could, only that we should be realistic about what it can achieve. I think the evidence is consistent with that view. But of course it is also consistent with the argument that the measures were very effective, and that the recovery would have been much weaker in the absence of these actions. And that is why the debate will continue.
In the meantime, there are some new areas of discussion emerging and, in a way, they are related.
Interest Rates and Savers
The first is the increasing concern that this world of ultra-low interest rates over a lengthy period is a big problem for savers. Here we are not talking about short-term trends. When everyone wants to save, the return to doing so will fall – that's economics. In a cyclical sense that has to be expected, just as costs of borrowing rise when demand is strong.
The issue is when long rates are very low for a long time. In such a world, the whole set of assumptions embodied in retirement income plans will be called into question. Increasingly, we hear commentary about the difficulty – or impossibility – of defined-benefit pension plans making good on their promises with long-term rates of return so low. The fact that accounting rules and regulation now strongly incentivise trustees to hold bonds – at the lowest rates of return in human history – so as to minimise mark-to-market valuation changes over the short term only exacerbates the problem.
The problem is surely not confined to defined-benefit plans. Accumulation arrangements are still predicated on some set of assumptions about future income needs and returns. It may take longer but surely many of the owners of these funds are going to feel disappointment. The implicit promises – even if made only to themselves – about their retirement incomes are in danger of not being fulfilled. It is not a very daring prediction to say that these issues will loom ever larger over the years ahead.
Some critics lay these problems at the door of the central banks, whose policy actions have worked to lower long-term yields on financial assets. If it were really true that central bank policies were the only factor at work in very low long-term interest rates, while at the same time they were not helping growth, the critics might have a point.
But are central banks alone responsible for the decline in long-term interest rates? Real interest rates have fallen noticeably since 2007 – nearly a decade ago now. For there to be persistent effects on real interest rates as a result of central bank actions is perhaps not impossible, but seems contrary to everything we were taught when we studied economics. Monetary policy is not supposed to be able to affect real variables – like real interest rates – on a sustained basis. Presumably, changes in risk appetite, subdued growth and expectations that growth will continue to be subdued have also played a role in lowering real rates.
Interest Rates and Growth
This brings into focus the really critical question: what are the prospects for sustained growth in the future? Relatedly, what expectations about rates of return in the future are reasonable? The real economy needs to generate decent returns on the real capital stock that are then matched (risk-adjusted) by the yield on financial assets. The financial assets are, in the end, just paper claims on that flow of real returns – directly in the case of private sector obligations and indirectly for government obligations, which rely on being able to tax growing private incomes. If the real economy can't perform to provide real returns to capital, there is nothing to back higher yields on financial assets. In that world, nominal and real yields on bonds would remain extremely low, the income being generated by those working with the capital stock would struggle to fund the benefits required by retirees through dividends and returns on bonds and bank deposits. Governments may not receive all the revenue they need to service their obligations. On the other hand, the stronger the prospects for long-term growth and good returns on the real capital stock, the smaller those problems will be and the more we can expect that, sooner or later, the yield on financial assets will be higher, in line with those real outcomes.
Which outcome will it be?
The more pessimistic are moving closer to the position of ‘secular stagnation’: that situation where the desire to save is so overwhelming and the apparent opportunities for profitable investment so weak, that real interest rates cannot equilibrate saving and investment for the system at positive rates of interest and full employment. The result is that the ex ante excess saving leads to a sustained below-full-employment equilibrium. The concept arose originally in the 1930s, but has recently been articulated by Lawrence Summers as a description of the current environment.
I still find this a bit too pessimistic, because I struggle to accept that today, to an extent virtually unprecedented in modern history, ingenuity, technological development, entrepreneurial drive and opportunity for improvement are so weak – so unprecedentedly weak – and people's desire to defer gratification so strong, that the equilibrium real rate of interest is actually going to be negative over an extended period.
What is undeniable, though, is that monetary policy alone hasn't been, and isn't, able to generate sustained growth to the extent people desire. Maybe this is simply the inevitable outcome after a period of excessive optimism and over‑leverage – an essentially cyclical explanation, where the cycle is a low‑frequency, financial one. Or maybe it is something more deep-seated and structural. That can all be debated. Either way, though, policies that encourage growth through means other than just ultra-cheap borrowing costs are surely needed.
It is often said, rightly, that policymakers should try to avoid unnecessary policy uncertainty. For central banks, this has meant trying to be clear about our objectives and our reaction functions – and what we will, or might, do in various states of the world. Maybe we need to be clearer about what we can't do. Monetary solutions are for monetary problems. If there are other problems in the underlying working of the economy, central banks won't be able to solve those.
It is this recognition that purely monetary actions can go only so far, coupled with the need for some more growth and more inflation, that lies behind the recent discussion of ‘helicopter money’. In essence, this approach, were it to be attempted, would really be fiscal policy or a combined fiscal-monetary operation. It could involve unrequited transfers (gifts) to individuals' bank accounts by the central bank – which diminishes the central bank's net worth and so would require the acquiescence of its owner. Alternatively, it could involve direct funding of government spending by central bank finance – monetary-fiscal coordination.
There would be a host of practical issues to sort out in the ‘helicopter money’ approach. Other commentators have talked about these recently.
The main complication is surely that it would be a lot easier to start doing helicopter money than to stop, if history is any guide. Governments have found that a difficult decision to get right. That is, after all, how we got to the point where direct central bank financing of governments is frowned upon, or actually contrary to statute, in so many countries. It would be a very large step to overturn those taboos, which exist for good reason. The governance requirements in doing so would be, if not intractable, at least very complex. Desperate times call for desperate measures, perhaps. Are we that desperate?
Before we even got close to that point, one would have thought that for many governments today there must still be projects of an infrastructure kind that would, through conventional fiscal operations at current bond rates, offer returns comfortably above their cost of funding. Helicopter money is surely not needed in these cases. Questions may arise, in some jurisdictions at least, in the minds of citizens about the ‘soundness’ of such conventional policies. But if such questions arise about conventional fiscal actions, it seems unlikely that adding central bank financing to the mix would allay them.
It all Comes Back to Growth
But the very fact – extraordinary as it is – that such possibilities are being openly discussed by serious commentators reinforces the point that, while people find global growth outcomes still a bit disappointing, we are reaching the limits of monetary policy in boosting it. Central banks must of course do what they can, consistent with their mandates, and they continue to explore options. It is certainly clever to find ways of pushing the effective lower bound for interest rates down a bit further. It is inventive to find ways of lending more, at more generous terms, to the private sector.
But surely diminishing returns are setting in. My suspicion is that more and more people realise this. Maybe this has something to do with market confidence being easily rattled. There was a hint in the recent episode of the feeling that central banks didn't have much left they could do, if things got worse.
So in the end we will collectively have to face up to the question of whether trend growth is lower and, if so, what is to be done about that. A few candidates might be advanced as contributing to such an outcome. Demographics is one. What we might label productivity pessimism – or is it realism? – along the lines of Robert Gordon's views might be another. Others would point to excess debt in many jurisdictions as another.
If trend growth is lower and we can't or don't want to do anything about that, then expectations about future incomes, tax bases and so on will have to be reconfigured. People will need some explanation of why we have to accept that outcome. It may be that this reconfiguration is, in fact, what is happening. That would help to explain why ultra-low interest rates are not, apparently, as successful in boosting growth in demand as might have been expected. The future income against which people would borrow looks lower than it did, not to mention that the current income against which some already had borrowed has turned out to be lower than assumed.
Alternatively, even if we accept that demographic headwinds and the legacy of earlier problems make growth harder to achieve, perhaps we can re-double our efforts to address the things that may be unnecessarily restraining growth today and in the future. They might be things like:
- in some jurisdictions, inadequately capitalised banks
- over-leveraged firms or households
- poor incentives for risk-taking of the ‘right’ kind
- practices that unnecessarily impede productivity, or that slow down the re‑allocation of capital from old industries to new ones.
If we could engender a reasonable sense that future income prospects are brighter, that there is a good return to innovation and ‘real economy’ risk taking, and so on, then people might use low-cost funding for more productive purposes than just bidding up the prices of existing assets. Over time, the return on financial instruments could rise in line with returns in the real economy. Pension funds and insurers would be better able to meet their obligations. Governments would more easily service their debts. Citizens, having had some explanations as to why changes were necessary, would, in time, see some gains in their way of life, or at least some threats to their living standards abate. They would see less resort to very unorthodox policies, because there would be less need for them. And I can't help thinking they would feel better about that.
We have lived through various bouts of financial market volatility before and doubtless these will recur from time to time. To some extent, this is inevitable.
But we can limit the damage from these mood swings by keeping a strong focus on improving growth fundamentals. It is surely time that policies beyond central bank actions did more in this regard. Our inability, so far, durably to lift growth prospects is arguably the biggest vulnerability the global financial system faces today. This needs to be our focus.
I thank Emily Perry for assistance in compiling these remarks. [*]
It's also important to recall that many markets never had that much real liquidity anyway. Outside the United States, corporate bond markets weren't really liquid at any time and some observers say that even in the United States – the biggest corporate market – genuine liquidity was always a bit doubtful. 
See Summers L (2016) ‘The Age of Secular Stagnation: What Is It and What To Do About It’, Foreign Affairs, 15 February 
See, for example, Bernanke (2016), ‘What tools does the Fed have left? Part 3: Helicopter money’, Brookings Institution, 11 April