Speech Some Longer-run Consequences of the Financial Crisis
Glenn Stevens [*]
Address to The Anika Foundation Luncheon
Supported by Australian Business Economists and Macquarie Bank
Thank you for coming along today in support of The Anika Foundation's work supporting research into adolescent depression. This is the fifth such occasion and it is very gratifying indeed to see such a strong response from the financial community. I want also to record my thanks to the Australian Business Economists for their support and to Macquarie Bank for their sponsorship of today's event.
My subject is the consequences of the financial crisis. We are all aware of the immediate and short-term impacts the crisis had on the international financial system and the world economy. I won't repeat them.
The initial phase of the recovery has been underway for over a year now. Global GDP started rising in mid 2009. When all the figures are in we will probably find that it rose by close to 5 per cent over the year to June 2010, though the pace has been uneven between regions and with some of the leading Asian economies seeking to slow down to a more sustainable pace, and European nations tightening fiscal policy, there is a bit more uncertainty just now about prospects for 2011. The bulk of financial institutions most affected by the crisis have returned to profit, while estimates of the total losses to be absorbed from the whole episode have tended to decline somewhat lately (though they are still very large). Financial market dislocation has gradually eased, albeit with sporadic episodes of renewed doubts and instability.
But what of the longer-run consequences of the crisis? I want to offer some remarks under three headings, though with no claim this is an exhaustive list. These remarks are about the general international situation, not Australia in particular, unless otherwise noted.
The first lasting consequence is the fiscal burden taken on by countries at the centre of, or close to, the crisis. There are three components to this.
First, some governments took on bank ownership in order to ensure the replenishment of capital that had been too thin to start with and that was depleted by the losses on securities and loans. Table 1 shows public capital injections to the financial sector for several key economies. The amounts in mainland Europe could quite possibly grow soon as a result of the forthcoming stress tests. Note that this is not necessarily a permanent burden since, if carried out successfully, the ownership stake can be sold again in due course. In fact about 70 per cent of the funds invested by the United States in banks have been repaid, and the US Government expects to make an overall profit from these capital injections. Nonetheless for a period of time governments are carrying a little more debt than otherwise as a result of the provision of support to the banking system.
|Capital Injections to
|Per cent of 2009 GDP||
Per cent of GDP
(a) Cumulative effect of fiscal stimulus from 2008–2010 as a per cent of 2008 GDP
Sources: Bloomberg; De Nederlandsche Bank; Eurostat; IMF; OECD; Thomson Reuters
Second, the depth of the downturn saw recourse to discretionary fiscal packages. As the table shows, while there was a lot of national variation, for some countries this spending was quite significant relative to the normal pace of annual growth in GDP. To the extent that the packages had measures that increased spending for a finite period but not permanently, the result is a rise in debt of a finite magnitude, but not an ever-escalating path of debt.
But it is the third factor – namely the magnitude of the downturn itself and the initial slowness of recovery – that is having by far the biggest effect on debt ratios. According to the IMF, for the group of advanced economies in the G-20, the ratio of public debt to GDP will rise by almost 40 percentage points from its 2008 level by 2015. Fiscal stimulus and financial support packages will account for about 12 percentage points of this. Close to 20 percentage points are accounted for by the effects of the recessions and sluggish recoveries. Another 7 percentage points comes from the unfavourable dynamics of economic growth rates being so much lower than interest rates for a couple of years.
Now it is somewhat inaccurate to attribute the economic downturn effects entirely to the financial crisis because there would probably have been some sort of slowdown even without a crisis. There will always be a business cycle, after all, and deficits and debt rise when downturns occur. As a comparison, the rise in the debt ratio of the G7 from 2000 to 2005 associated with the previous cyclical downturn – which was not an especially deep one – was around 12 percentage points.
Nonetheless the recent downturn was a bad one in many countries, and that is because it was associated with a financial crisis. For this reason, together with the other factors I have already mentioned, the major countries generally are going to have significantly higher public debt relative to GDP after the crisis than before, and the debt ratios will continue to rise for several more years.
This was largely unavoidable. To a considerable extent, the fiscal legacy can be seen as one manifestation of a broader legacy of lost output (and hence weaker budgetary positions through ‘automatic stabilisers’) over a period of several years. Generally speaking, the public balance sheet has played the role of a temporary shock absorber as private balance sheets contracted.
But the servicing of the resulting debt is an ongoing cost to the citizens of the countries concerned. At present that additional cost is, in some countries, reduced compared with what it might have been due to the low level of interest rates on government debt that we see. Moreover had the debt not been taken on it could well be that the economic outcomes would have been much worse, so increasing fiscal and other costs. Nonetheless this lasting debt servicing burden is a real cost.
More importantly, the pace of the rise in public debt has increased focus on the question of fiscal sustainability. This is especially so in those countries where debt burdens were already considerable before the crisis.
The difficulty is that ‘sustainability’ is so hard to assess. It is more complex than simply the ratio of debt to GDP. In any number of countries, including our own, public debt ratios have on some past occasions been much higher than 100 per cent. Many countries found themselves with such a situation in the aftermath of World War II. Those ratios thereafter came down steadily though it took until the 1960s in our case, or longer in some others, for them to reach levels like 50 or 60 per cent that today is often regarded as a sort of benchmark. That reduction occurred for a combination of reasons. The big deficits of the war years really were temporary in most cases; economies recorded good average rates of output growth in the long post-war boom with strong growth in both population and productivity; in the same period, business cycle downturns were not especially deep or protracted; interest rates were low – so the comparison of the growth rate of GDP and the interest rate on the debt was favourable; and lastly, significant inflation raised the denominator of the ratio – in some cases in the late 1940s and early 1950s, and more widely in the 1970s.
So high or even very high debt ratios per se have not necessarily been an insurmountable problem in the past. On the other hand, that earlier decline in debt ratios may not be easy to replicate in the future. In some countries demographics are working the wrong way, with population growing more slowly or even declining. Other things equal, future growth of nominal GDP will thus be lower than in the past. A period of rapid catch-up growth in income, which helped Europe and Japan in the couple of decades after 1950, is more likely in the future to occur in the emerging world than in the parts of the developed world where most of the debt is.
In fact it might be argued that the fiscal position of a number of countries has been increasingly vulnerable for quite some years. Perhaps what the crisis has done is to act as a catalyst to bring forward a set of pressures for long-term budgetary reform that were bound to emerge anyway.
This has placed some governments in a very difficult bind, since the heightened focus on sustainability has increased the pressure for fiscal consolidation at a time when aggregate demand remains weak. The ‘least-damage path’ through the various competing concerns has become harder to tread.
Public Intervention in Finance
The second long-run implication of the crisis is that government intervention in the financial sector has become much more pervasive. I have already mentioned governments taking major stakes in banks in key countries, which was virtually unthinkable, certainly for an American or British government, only three years ago
But the intervention was broader than just a temporary period of public ownership – as massive an event as that has been. Take guarantees. Once the Irish Government guaranteed its banks, governments all over the world felt bound to follow suit in some form or other – expanding or (as in our case) introducing deposit insurance, and guaranteeing wholesale obligations (for a fee). The feeling was probably most acute in countries whose citizens could shift funds to a bank guaranteed by a neighbouring country without much effort.
In circumstances of incipient or actual panic, or potential complete market closure, measures along such lines had to be taken. The simple truth is that, given a big enough shock, the public backstop to the financial system has to be used.
But the backstop having been used so forcefully on this occasion, it is desirable not to use it again soon. The real question is how, having set the precedent, governments avoid too easy recourse to such measures in the future. They will want to get to a position where in future periods of financial turmoil, they are standing well in the background, not in the foreground.
Meanwhile there is a growing debate, at a very high level, about what the financial sector should do, and what it should not do. The number of inquiries, commissions, conferences, papers and ideas about the desirable shape of the system in the future is growing. This is a growth industry with, I should think, pretty good prospects over the next few years.
Another characteristic of public intervention is the expansion of central bank balance sheets. During a panic, the central bank's job is to be prepared to liquefy quality assets, with a suitable combination of hair-cuts and penalty rates, to the extent necessary to meet the demand for cash.
Once the panic is over, the additional liquidity shouldn't need to remain in place, and indeed some particular facilities established by central banks had design features which saw their usage automatically decline as conditions improved. But overall it has proven difficult, so far, for the major central banks to start the process of winding down the sizes of their balance sheets.
In effect central banks have been replacing markets. They had to. If counterparties feel they cannot trust each other and flows between them are cut off, with everyone preferring to keep large liquid balances with the central bank, the central bank has to replace the market to ensure that everyone has the cash they need each day (against suitable collateral of course). Central bank purchases have also acted to reduce credit spreads and yields.
I am not arguing that this policy is macro-economically wrong. But consider the implications of persisting with it over a long period. One doesn't have to believe that markets can solve all problems to accept that well-functioning markets have a value. A cost of the zero or near zero interest rate and a greatly expanded role for the central bank's balance sheet is that some markets tend to atrophy – as Japan has found over a decade.
Moreover some central banks have had to accept a degree of risk on their own balance sheets that is considerably larger than historical norms. Of course since the governments are ultimately the owners of the central banks, that is where the risk really resides. From a purely financial point of view, the risk of a rise in yields on bonds held by central banks, but issued by their own governments, is actually no risk at all once the central bank is consolidated with the government. On the other hand, to the extent that central banks are really exposed, or are exposing their governments, to private credit risk or to the risk of other sovereigns, those are genuine risks.
So some central banks, like their governments, have found themselves in very unusual terrain. It is terrain: in which the relationship between the central bank and the government is subtly changed; where the distinction between fiscal and monetary policy is less clear; from which it may be hard to exit in the near term; and a side effect of which may be wastage, over time, in some elements of market capability.
Of course I have not yet mentioned the other significant public intervention in finance which is the major regulatory agenda being pursued by the international community. This is being pushed by the G-20 process and by the Financial Stability Board. The ‘perimeter’ of regulation is being extended to include hedge funds and rating agencies. Governments are demanding a say in the pay of bankers and talking of specific taxes on banks' activities. The climate is more difficult for bankers these days, it seems, especially in countries where the public purse had to be used to save banks.
But the core work on regulatory reform is being done by the Basel Committee on Banking Supervision. I am not sure who first began to talk of this as ‘Basel III’ but the label seems to be starting to stick. Basel II came in only about two years ago for many countries, 20 years after Basel I. The gap between Basel II and Basel III looks like being a lot shorter. Warning: that pace of acceleration in devising new standards is unsustainable!
You would all be well aware of the essence of the proposals. In a nutshell, what regulators are pushing toward is a global banking system characterised by more capital and lower leverage, bigger holdings of liquid assets and undertaking less maturity transformation. It is hoped that this system will display greater resilience to adverse developments than the one that grew up during the 1990s and 2000s.
What will be the implications of the various changes? Put simply, the customers of banks around the world, and especially of large internationally active banks, will generally be paying more for intermediation services, in the form of higher spreads between rates paid by banks and rates charged by them. The reason is that capital is not free and it typically costs more than debt. The spread between a bank's own cost of debt, both deposits and bonds etc, and the rate it charges its borrowers has to cover operating costs, expected credit and other losses and the required cost of equity capital. Assuming the costs of equity and debt do not change, the more capital intensive the financial structure is, the higher that spread has to be. A requirement to hold more high quality liquid assets and/or to lengthen the maturity of debt has a similar effect.
Of course the costs of equity and debt may not be, and actually should not be, constant as banking leverage declines. The cost of wholesale debt should fall over time if the equity buffer, which protects unsecured creditors against losses, is larger. In time, the cost of equity may even fall with lower leverage if the required equity risk premium declines to reflect a less variable flow of returns to equity holders. All of that assumes of course that the perceived riskiness of the underlying assets is unchanged.
Still, such effects would take some time to emerge. Most observers appear to agree that even allowing for some possible pricing changes over time, spreads between banks' borrowing and lending rates will be wider in the new equilibrium after the regulatory changes have been fully implemented.
What will be the broader economic effects of these higher costs of intermediation?
The conclusion most people are reaching is that economic activity will, to some extent and over some horizon, be lower than otherwise. The question is, by how much and for how long? There are various ways of approaching that question. Researchers are putting it to various macroeconomic models. The answers will vary, depending on the models and particularly according to the degree of detail in models' financial sectors. Overall, these techniques are likely to show moderate but nonetheless non-zero effects on economic activity of the regulatory changes over an adjustment period of several years.
Some other analyses, often by banks themselves, find much larger adverse effects. This is usually because they find that credit to the private sector must be reduced in order to meet the various standards, particularly liquidity standards, because it is assumed there will be quantity limits on the availability of funding in the form necessary. It is further assumed that a mechanical relationship between credit and GDP exists, which in turn results in big adverse impacts on GDP.
To a fair extent these differences come down to a discussion about what economists would call elasticities: for the non-bank private sector to respond to a desire for banks to be funded differently, how big a change in the price is required – a little, or a lot? Some of the industry estimates appear (to me anyway) to assume elasticity pessimism. Official sector estimates are likely to be based on less pessimism.
In truth, it is impossible to know for sure exactly how big these effects will be. That is a reason to proceed carefully, and to allow time for the new rules to be phased in. Clearly, we wish the new rules to be constraining risk taking and leverage as the next boom approaches its peak, but that will probably be some years away, so we have time to implement strong standards and allow an appropriate period of transition.
That said, there are three broad observations that I would like to offer.
First, I think we ought to be wary of the assumption of a mechanical relationship between credit and GDP. Of course a sudden serious impairment in lenders' ability to extend credit almost certainly amounts to a negative shock for growth in the short term. But did the steady rise in leverage over many years actually help growth by all that much? Some would argue that its biggest effects were to help asset values rise, and to increase risk in the banking system, without doing all that much for growth and certainly not much for the sustainability of growth in major countries. Some gradual decline in the ratio of credit to GDP over a number of years, relative to some (unobservable) baseline, without large scale losses in output may be difficult to achieve but I don't think we should assume it is impossible.
Secondly, even accepting that there will probably be some effect of the reforms in lowering growth over some period of time, relative to baseline, we have to remember that there is a potential benefit on offer too: a global financial system that is more stable and therefore less likely to be a source of adverse shocks to the global economy in the future. So we have a cost-benefit calculation to make. Quantifying all this is very difficult, but then that is often the case when deciding policies.
Thirdly, however, the reforms do need to be carefully calibrated with an eye to potential unintended consequences. One such consequence, obviously, would be unnecessarily to crimp growth if the reforms are not well designed and/or implementation not well handled.
Another could be that very restrictive regulation on one part of the financial sector could easily result in some activities migrating to the unregulated or less regulated parts of the system. Financiers will be very inventive in working out how to do this. If the general market conditions are conducive to risk taking and rising leverage (which, sooner or later, they will be if the cost of short-term money remains at zero), people will ultimately find a way to do it. Of course while ever the unregulated or less-regulated entities could be allowed to fail without endangering the financial system or the economy, caveat emptor could apply and we could view this tendency simply as lessening any undue cost to the economy of stronger regulation of banks. But if such behaviour went on long enough, and the exposures in the unregulated sector grew large enough, policymakers could, at some point, once again face difficult choices.
The financial turbulence we have lived through over recent years has had profound effects. The most dramatic ones in the short-term have been all too apparent. But big events echo for many years. My argument today has been that the full ramifications are still in train, insofar as impacts on governments' finances, governments' role in the financial sector and the trend in regulation are concerned. It will be important, as these reverberations continue, for there to be a balanced approach blending strong commitment to sensible long-run principles with pragmatism in implementation.
In Australia we have been spared the worst impacts of serious economic recession in terms of lost jobs, much as we will be spared the prospect of higher taxes that face so many in the developed world. These are factors that support our native optimism, at least about economic conditions.
Nonetheless depression still ranks as a serious, and underestimated, problem in our community including among our young people. That is why the work of the Anika Foundation, working alongside other bodies seeking to combat depression, is so important, and why I thank you all very much for coming along today.
I thank George Gardner for assistance in preparing this address. [*]
Of course the United States retains the stake in the insurer AIG. Fannie Mae and Freddie Mac also remain in government ownership though that perhaps might be seen simply as final recognition on the US Government's balance sheet of an obligation everyone always assumed it would meet. 
A public debt to GDP ratio of 60 per cent was one criterion in assessing eligibility for the European monetary union and was a benchmark in the Stability and Growth Pact (a Pact perhaps more often honoured in the breach). 
By the way, in those countries that choose to impose ‘levies’ of some kind or other on banks, we shouldn't assume that the banks' shareholders will ultimately bear such costs: it is fairly likely that the costs of this tax will fall mainly on the customers.