Responsible long-term fiscal policy (pilot survey)

Wednesday, April 2, 2014

First question:

To help ensure that advanced country governments have sufficient flexibility to respond to future crises, it is important that finance ministries aim for a ratio of public debt to GDP that is substantially less than 60% in normal times.

Second question:

To help ensure that advanced country governments pursue responsible fiscal policies, countries should adopt formal rules for limiting structural deficits, which are supported by primary legislation or constitutional reform.

Background

This pilot of the CFM survey focuses on an important but difficult question, namely whether more discipline should be imposed on long-term fiscal policy. The first question focuses on the ratio of public debt to GDP and the second question focuses on the deficit.

Since the outbreak of the financial crisis, the ratio of public debt to GDP has sharply increased in most developed countries and the yield on government debt increased substantially in many of these countries and may become particularly volatile as policy rates normalise. High public debt to GDP ratios and the scope for increased costs of borrowing may place considerable constraints on the future ability of governments to pursue stabilization policies to stimulate the economy.

Advanced countries may yet fully recover from the recent financial crisis nevertheless, there are likely to be other episodes in which government liabilities may increase sharply. Moreover, government liabilities may increase as the baby boomer generation leaves the work force. Before the recent financial crisis, a ratio of sovereign debt to GDP below 60% was considered unproblematic for developed countries.

The first survey question asks whether governments should target much lower ratios in normal times. The question asks whether the target ratio should go down, but is silent on whether the target should be for gross debt or for net debt. The advantage of using gross debt is that reported numbers are less easy to manipulate whereas the advantage of using net debt is that it better represents the total burden carried by the government sector.

The second survey question asks whether constitutional amendments should be imposed to restrict the size of the fiscal deficit. Several developed countries are considering or have already approved amendments to their constitutions to introduce constraints on the scope to run discretionary fiscal policy. The German constitution has been amended to introduce the "Schuldenbremse" (debt brake), which will limit the structural deficit (0.35% of GDP from 2016 onwards and no structural deficit after 2020). Similarly, a balanced budget amendment to the constitution has been approved in Slovenia and Spain. Several other countries, such as Austria and Italy are considering constitutional amendments to restrict fiscal policy. In the U.S., such amendments to the constitution have frequently been proposed, although none have yet been passed. The second question does not offer a method for measuring the structural deficit and so each panel member is asked to answer this question under the assumption that structural deficits are measured in a feasible manner, which they may outline if they wish.

What did the experts say?

Question 1: Lower debt to GDP ratio in normal times?

53% of the panel members agree or strongly agree that the government should target lower ratios and 42% disagree or strongly disagree. If the answers are weighted with confidence levels, then these outcomes are very similar. Those that are in favor seem a bit less explicit about the motivation, but the arguments that are given seem to focus on being prudent and having more room for responding when bad times hit. Those that (strongly) disagree point out that the debt to GDP ratio is only one particular indicator of fiscal policy and a crude measure. Moreover, it is pointed out that some countries can probably function well with high debt to GDP ratios. Moreover, it is pointed out that generally any target is likely to be different for different countries.

Question 2: Constitutional amendment?

The answers are even more equally split than for the first question. Those in favor emphasize the benefit of a sustainable long-term position and those against point out that there are times when the lack of flexibility can be very costly. It is possible that not all panel members interpreted the question the same. For example, some panel members pointed out that it is important to allow for high deficits during emergencies. The question was silent on whether the constitutional amendment would allow for this.

Several panel members (Ethan Ilzetzki , Helene Rey, Richard Portes, and Coen Teulings) suggest independent fiscal committees "with teeth" and a "larger role for expert opinion", that is, getting closer to how monetary policy is conducted. It would be interesting to see whether the panel members would be more united on the creation of such committees.

How the experts responded

What is the right sovereign debt to GDP ratio in normal times?

Participant	Answer	Confidence level	Comment
Jonathan Portes KIng's College, London	Agree	Confident	Yes but A) there is no threshold value (60,90, whatever) over which debt is dangerous B) In immediate aftermath of crises (like now) priority should be for fiscal policy to support recovery
Jan Eeckhout University College London	Agree	Very confident
Alan Sutherland University of St. Andrews	Agree	Very confident
Nicholas Oulton London School of Economics	Strongly Agree	Very confident	In the UK the debt-GDP ratio averaged about 40% pre-crisis but this turned out to be too low to leave adequate room for responding to the crisis.
Paul De Grauwe London School of Economics	Disagree	Confident
Kevin Daly Goldman Sachs	Agree	Confident
John VanReenen London School of Economics	Strongly Disagree	Very confident
Andrew Mountford Royal Holloway	Strongly Disagree	Extremely confident
Morten Ravn University College London	Agree	Confident	Some target for debt / GDP may be useful because it helps building credibility and avoid debt getting "unsustainable". But optimal debt levels will almost certainly vary across countries (due to structural differences) and across time so I am not sure that 60 percent is the right value.
Giancarlo Corsetti University of Cambridge	Disagree	Very confident
Helene Rey London Business School	Strongly Disagree	Confident	High debt to GDP ratios can and have been sustained by advanced economies. The crisis has proved once more than in some cases counter-cyclical fiscal policies are key macroeconomic policy tools that should be used. For countries participating in a currency union, a common fiscal policy framework has to be agreed on but it should not be the one proposed here. Finally on gross debt versus net debt: one should worry about multiple equilibria for very high levels of debt (way above the thresholds discussed here). For multiple equilibria and runs, it is gross debt that matters and the maturity structure as well.
Richard Portes London Business School and CEPR	Strongly Disagree	Very confident	Recent empirical work suggests there are no thresholds beyond which the debt-to-GDP ratio is associated with lower economic growth. History shows that economies can cope with very high ratios, although it may take an extended period to return to 'normality'. But 'normal' may encompass a very wide range. The fixation on debt reduction since the crisis has been bad economic policy, with demonstrably bad results.
Romesh Vaitilingam Royal Economic Society	Agree	Confident
Fabien Postel-Vinay University College London	Agree	Confident
Wendy Carlin University College London	Disagree	Confident	Naming a target is an aspiration that is not very meaningful. Moreover it provides no guidance on the transition from where the economy is now to such a target. Better is to focus attention on the elements of a prudent fiscal policy, which ensures long run sustainability of the public finances, and allows for stabilization when necessary.
Sushil Wadhwani Wadhwani Asset Management	Disagree	Confident
Coen Teulings University of Cambridge	Disagree	Very confident	A debt-to-GDP ratio is to crude a measure to be useful for policy making. Spain had a lower ratio than Germany, and nevertheless Spain is in trouble, not Germany. The reason is that the sustainability of sovereign dept depends on many other factors, inter alia the growht rate of country (Italy), the balance sheet position of the private sector, the availability of private pension funds (Netherlands, Swiss), implicit PAYG liabilities of the government, market distortions leading to rents of which the net present value is tradable on financial markets, etc.
Angus Armstrong Rebuilding Macroeconomics, IGP, UCL	Agree	Confident	I would prefer gross debt to be used as it is easier to compare like with like in cross country applied work.
David Cobham Heriot Watt University	Agree	Confident	My view is that a ratio of around 40% (gross) would be an appropriate target, but it is important to say that this should be attained gradually over a period of 'normal times', and the target should not be used as an excuse for more austerity now.
Tim Besley London School of Economics	Strongly Agree	Very confident
John Driffill Birkbeck College, University of London	Disagree	Confident	60 percent may be ok for countries with an independent monetary policy, like the UK, US and Japan, but too high for Euro zone members without the same flexibility, for whom a much lower limit such as 30 % might be needed
Christopher Martin University of Bath	Neither agree nor disagree	Confident	Any stable ratio less than 100% wold be fine
Marco Bassetto University College London	Disagree	Confident	30% or 60% does not make a huge difference in the face of catastrophic shocks. For other types of shocks, 60% probably leaves enough of a cushion; what is more important, is to have a credible commitment mechanism to restore fiscal balance in a reasonable time after a shock hits.
Tony Yates University of Birmingham	Agree	Confident	60% was probably a good guess before the crisis. Since the crisis, we have upped our estimate of the frequency and cost of crises. I don't personally hold out that much hope that regulatory reform enacted will do a whole lot of good. That leaves me thinking we need to make more fiscal room for possible future bailouts, and extreme automatic stabiliser responses. However, we aren't going to be able to start down the road of getting to <60% any time soon.
Kate Barker British Coal Staff Superannuation Scheme and University Superannuation Scheme	Agree	Confident
Sir Christopher... London School of Economics	Disagree	Confident	A lower debt than the 60% threshold would not be bad but the transition to it will be bad. So although I support zero deficits in good times (which would reduce the debt/GDP ratio) I do not support large surpluses that will bring the ratio of high-debtors below 60%
Costas Milas University of Liverpool	Agree	Confident
Wouter Den Haan London School of Economics	Strongly Agree	Very confident	In Iceland, Ireland and Spain, low initial debt to GDP ratios did not prevent debt to GDP ratios reaching problematic levels. Without such low initial debt to GDP ratios, however, these countries probably would have faced larger and more persistent problems. Moreover, if all countries in the Euro area would have started out with a debt to GDP ratio below 30%, then the Euro zone debt crisis would not have been so severe and possibly would have been over by now.
Andrew Scott London Business School	Strongly Disagree	Extremely confident
Ethan Ilzetzki London School of Economics	Agree	Very confident	I do not think 60% is a magic number and in fact might be too high for "normal times", but better a target than none.
Gianluca Benigno London School of Economics	Disagree	Confident
Paolo Surico London Business School	Neither agree nor disagree	Confident	An explicit reference value has proved a very transparent strategy to communicate monetary policy actions and a very effective guidance to coordinate the public expectations under (de-iure or de facto) inflation targeting regimes. It seems plausible to think that a similar framework may be useful to anchor investors' expectations in the government bond market. On the other hand, it seems hard to justify a specific numerical value relative to another one, as much as it is not unconceivable that an inflation target of, say, 3% (as opposed to the actual 2% in the UK) would have achieved a similar performance in terms of inflation stability since 1992.
Michael McMahon University of Oxford	Disagree	Very confident	I do not agree that 60% or some other lower value is somehow a magic number that, if adhered to, will provide the scope for a fiscal response in the event of another crisis or even a milder recession. The application of one-size-fits-all optimal level of debt does not seem like good advice to me. We know, in an accounting sense, that debt dynamics (in terms of debt as a percentage of GDP) are affected by nominal GDP growth which differs substantially across countries. Countries face very different future claims from health and pension spending, and moreover governments in different countries have different levels of assets. It may be that certain countries need to reduce debt to levels substantially below 60%, but also there are some countries who could, and in fact should, allow debt to be higher than 60% in order to facilitate investment in the near term.
Jagjit Chadha National Institute of Economic and Social Research	Strongly Agree	Very confident	Nervous public debt in normal times should be under 40% of GDP.
Michael Wickens Cardiff Business School & University of York	Agree	Confident	The level of the debt-GDP ratio should not be considered on its own. Many countries have survived for years with ratios over 100% - eg Belgium. The real question is whether the fiscal stance can sustain any level without running the risk of a credit downgrade which would raise the cost of borrowing.
David Smith Sunday Times	Agree	Very confident	The experience of the crisis was that even countries with low debt to GDP ratios, such as the Uk and New Zealand, saw them rise very sharply. It is prudent to aim low.

Should formal rules be used to limit structural deficits?

Participant	Answer	Confidence level	Comment
Jan Eeckhout University College London	Agree	Very confident
Giancarlo Corsetti University of Cambridge	Agree	Very confident
Kevin Daly Goldman Sachs	Disagree	Confident
Sir Christopher... London School of Economics	Disagree	Confident	Having formal rules prohibiting structural deficits would be too rigid. I support balanced structural budgets but given the difficulty of defining the structural deficit and the need for more flexibility sometimes I am against making this a legal requirement
Costas Milas University of Liverpool	Strongly Agree	Very confident
Morten Ravn University College London	Agree	Very confident	Rules are a useful for building a sustainable long term position. A country like Sweden that has a surplus target did relatively well during the recent crisis. Rules have little credibility unless they are formal so some kind of legislation may be useful. It may also be useful then to state circumstances under which the rules can be overcome or temporarily adjusted.
Helene Rey London Business School	Disagree	Confident	I would favour independent fiscal councils with teeth rather than rules based on a concept (structural deficit) which is bound to be badly measured. At the European level, there should be a body coordinating the work of the national fiscal councils.
Richard Portes London Business School and CEPR	Strongly Disagree	Very confident	'Balanced-budget amendments' may have some justification for regional sub-units within a sovereign (states, municipalities). But they impose undesirable inflexibility on economic policies for sovereigns which may have to respond to severe aggregate macroeconomic shocks. A much better alternative is an independent fiscal commission that analyses and comments on government fiscal policies.
Romesh Vaitilingam Royal Economic Society	Agree	Confident
Fabien Postel-Vinay University College London	Strongly Agree	Confident
Wendy Carlin University College London	Disagree	Confident	Unless crises of the kind recently experienced can be ruled out, this is not a sensible approach. A common approach to balanced budget rules across countries with different exchange rate regimes and with different wage setting institutions (which may provide scope for internal devaluation as a substitute for fiscal policy in stabilization) is not appropriate.
Sushil Wadhwani Wadhwani Asset Management	Strongly Disagree	Very confident
Coen Teulings University of Cambridge	Agree	Not confident	Issue on the timing of fiscal policy and the provision of intergenerational Insurance (both are closely related) are complicated and too hard to explain to voters and to have a policy debate on. Hence, voters in developed countries have a preference for austere politicians, since they are highly suspicious about politicians who want to spend. This leads to excess austerity in bad times and too little austerity in good times. One might therefore contemplate in a larger role for expert opinion on fiscal policy, similar to what is more and more common practice in monetary policy.
Angus Armstrong Rebuilding Macroeconomics, IGP, UCL	Disagree	Very confident
Alan Sutherland University of St. Andrews	Agree	Very confident
Simon Wren-Lewis University of Oxford	Agree	Very confident	Rules should be supported by independent fiscal institutions, which can advise the public if the rules are be adhered to, and when it might be sensible to break them.
David Cobham Heriot Watt University	Neither agree nor disagree	Confident	I have two reasons for neither agreeing nor disagreeing. The first is that the measurement of structural deficits is so difficult that it is open to abuse as well as innocent errors, and the latter can be enormous - e.g. the OECD's estimates in successive Economic Outlooks for the UK's structural deficit in 2007 have risen from below 0.5% (2007 estimate) to 4.9% (2013estimate). Second, trying to incorporate such limits in legislation is partly petty politics (trying to provoke the opposition) and partly dubious economic policy (trying to fix policy for all future time, in a way that ignores possible shocks and trends). It is better to argue out the need for appropriate constraints in a democratic way, which means convincing and reconvincing successive generations of politicians and electors, and accepting that in an uncertain world governments just have to allowed to exercise at least some discretion.
Tim Besley London School of Economics	Disagree	Very confident
John Driffill Birkbeck College, University of London	Agree	Confident
Jonathan Portes KIng's College, London	Strongly Agree	Confident	Whether it is sensible to target the absolute or structural deficit may well depend on the political circumstances - a country with a record of responsible fiscal policy may be in a better position to adopt a more gradual approach.
Christopher Martin University of Bath	Disagree	Very confident	Fiscal policy is not an appropriate tool for stabilisation. The "structural" component of a deficit is not useful in practice
Paul De Grauwe London School of Economics	Disagree	Confident
Marco Bassetto University College London	Agree	Not confident	Some countries have managed their debt very well without resorting to such measures. Others would most likely have been better served by sharper commitment mechanisms. The challenge is to build escape clauses that do not unreasonably constrain governments in unforeseen circumstances but at the same time do not empty the formal rules of any bite; perhaps this can be done requiring supermajorities to trigger the escape clause, but I am not confident that this would be sufficient for the rules to retain their bite. For countries that share a monetary union, the recent experience suggests that there are strong incentives to provide bailouts, and hence the case for enforceable multilateral agreements is much stronger.
Tony Yates University of Birmingham	Neither agree nor disagree	Confident	Seems odd to have confidence level and level of agreement. Anyway, I'm not convinced that such rules would leave enough freedom to deal with catastrophes, nor be workable, since the distinction between structural and cyclical is ambiguous, and could never be made both legally watertight and economically sensible. [Eg, we could tie the constitution to the hp filter of the deficit, but that would be daft.]
Kate Barker British Coal Staff Superannuation Scheme and University Superannuation Scheme	Neither agree nor disagree	Not confident
Francesco Caselli London School of Economics	Disagree	Confident
Nicholas Oulton London School of Economics	Strongly Agree	Very confident	I strongly agree with the proposition though I am no nearly so confident that enforceable rules can be put in place. It is very difficult to stop people making promises when they won't be around to be held to account when the promises come due. This is true in the private as well as the public sector.
Wouter Den Haan London School of Economics	Neither agree nor disagree	Confident	The answer is not the same for all countries. "Rules" that limit structural deficits make it possible to be in a strong financial position when crises hit. Constitutional amendments allow for little or no flexibility to respond to unforeseen events. This is an important drawback. In some countries, however, constitutional amendments may be necessary to impose credible discipline.
Andrew Mountford Royal Holloway	Strongly Disagree	Extremely confident
John VanReenen London School of Economics	Strongly Disagree	Extremely confident
Andrew Scott London Business School	Agree	Not confident
Ethan Ilzetzki London School of Economics	Agree	Not confident	I would like to see institutional changes to give less discretion to finance ministries to run deficits. My lack of confidence stems from the fact that I am not sure that a formal rule is the best way to do so. We have seen the limits to the applicability of monetary rules in this recession and a rigid rule might pose problems in unusual circumstances. Fiscal committees might be an alternative way to impose fiscal discipline.
Patrick Minford Cardiff Business School	Disagree	Confident	Budget deficits are needed in times of emergency. Yet commitment to responsible fiscal behaviour (ie at present to getting debt down again as % of GDP) is also necessary. Constitutional rules will simply obstruct the former while not adding to commitment; indeed their necessary breaking would create noise. I prefer the current experiment with a 'Fiscal Advisory Authority in the UK ('Office of Budget Responsibility'). Here commitment is maintained via the need to demonstrate soundness; but ability to respond to events is preserved.
Gianluca Benigno London School of Economics	Agree	Confident
Paolo Surico London Business School	Strongly Agree	Confident	There is a classical trade-off here between rules and discretion and the hard judgement call is about the size of the maximum structural deficit (in percentage of GDP) that minimizes the losses associated with each scenario. While I am in favour of a numerical value (see also my comment above), I find the proposal of running a balanced budget too restrictive and probably less credible than a lower percentage of GDP.
Michael McMahon University of Oxford	Agree	Confident	I am in favour of rules governing fiscal policy limits so long as those rules ensure that elected politicians can select the balance and allocation of spending and tax within those limits. Any legislation must set rules that take account of different country situations, and expected future liabilities; this means that structural deficit limits may need to be time-varying. And it is vital to have an appropriate system for adjusting appropriately for cyclical effects; I remember that an attempt by the European Commission to cyclically-adjust EU fiscal deficits failed because countries could not all agree on a single method to cyclically-adjust the deficit.
Luis Garicano London School of Economics	Agree	Confident
Jagjit Chadha National Institute of Economic and Social Research	Disagree	Very confident	Need to have flexibility over deficits.
Michael Wickens Cardiff Business School & University of York	Disagree	Very confident	In the euro zone when monetary policy is too loose - eg for a high inflation country - it is necessary to have a tighter fiscal policy than if inflation is close to the euro zone average.
David Smith Sunday Times	Disagree	Confident	There will be times when governments need to adopt discretionary fiscal policy, even when such action would result in a rise in the structural deficit. Such rules are likely to be too limiting.
Silvana Tenreyro London School of Economics	Agree	Confident	I agree with the principle, but not necessarily with the tight margins in Germany or the rather extreme balanced-budget requirements in Slovenia.

How the experts responded

What is the right sovereign debt to GDP ratio in normal times?

Should formal rules be used to limit structural deficits?

About the CFM Surveys