Commentary: the government's fiscal strategy.

Portes, Jonathan

This Commentary examines the government's fiscal plans for the 2015-20 Parliament, and theoretical and practical issues raised by the new fiscal framework. It does not analyse the short-term macroeconomic impacts of the changes to the fiscal stance, which are discussed in the chapter on the prospects for the UK economy elsewhere in this Review.

Changes to fiscal plans for this Parliament

The most notable point about the government's fiscal plans, relative to those published in the March Budget, is that the profile for spending now looks both more realistic and more sensible. Many commentators had pointed out that the March plans implied extremely and frankly implausible--sharp cuts in current spending on public services in 2016-17 and 2017-18, followed by some rises in spending in subsequent years. There was never any economic rationale for this profile, described at the time by the Office of Budget Responsibility (OBR) as a "rollercoaster profile" (OBR, 2015a). The OBR now notes (OBR, 2015b):

"The new Government has used its first Budget to loosen significantly the impending squeeze on public services spending that had been pencilled in by the Coalition in March. This is being financed by welfare cuts, net tax increases and three years of higher government borrowing ... The Government has delayed the expected return to a budget surplus by a year to 2019-20, but is then aiming for a slightly bigger surplus in the medium term ... On the basis of these provisional plans, the forthcoming Spending Review would be a lot less challenging than it appeared in March."

As the OBR points out, this represents a significant further slowing in the pace of fiscal consolidation.

This reprofiling appears entirely sensible from both a macroeconomic perspective--there is no strong macroeconomic rationale for front-loading spending cuts in order to achieve an earlier surplus--and from an implementation one. In practical terms, very few analysts thought that the projected path of public service spending set out in the March Budget could be delivered without risking serious damage to some key public services.

[FIGURE 1 OMITTED]

But although it no longer looks impossible, that is not to say that the spending profile in this Budget will be easy to deliver; the Spending Review later this year will have to confront some very difficult tradeoffs. Even in protected areas, such as health spending, plans look at best barely adequate to meet growing demands; while some unprotected areas (policing, prisons, business support) will face continued very sharp cuts (IFS, 2015). Some indication of the scale of the cuts required was given by the Budget measures that will have the effect of sharply increasing the cost of university education for poorer students.

One last point worth noting is that, during the last Parliament, the government justified its failure to meet its fiscal targets by noting that it had largely (although by no means entirely) stuck to its spending plans; the shortfall was largely driven by shortfalls in tax revenues, both cyclical and structural. There was no 'Plan B', the Chancellor argued in 2013, because "government spending cuts have continued as planned", despite the serial failure to hit borrowing projections (Osborne, 2013). This was entirely sensible: at the time, I welcomed the fact that the government had not attempted to hit its targets by additional spending cuts or tax increases, while noting the inconsistencies in this argument (Portes, 2014).

Now, however, in this Budget, the OBR has revised up its forecasts for tax revenues quite significantly over the forecast period. Consistency would appear to require that this should feed through into lower borrowing rather than higher spending: as the above OBR summary notes, the opposite is the case (the Budget increases both). So, just as in the last Parliament, but from an opposite perspective and with an opposite rationale, the government has reacted to exogenous economic developments to slow the pace of fiscal tightening. Again, this is welcome from a macroeconomic perspective. However, it exemplifies a continued willingness on the part of the Chancellor to pick and choose between different rationales and targets in an inconsistent and at best confusing way; this does not necessarily bode well for the operation of the new fiscal targets over the longer term.

The new fiscal rules

Alongside the Budget, the Treasury published a new draft of the Charter for Budget Responsibility (HM Treasury, 2015a). The draft Charter sets out the Treasury's objectives for fiscal policy:

* ensure sustainable public finances that support confidence in the economy, promote intergenerational fairness, and ensure the effectiveness of wider government policy;

* support and improve the effectiveness of monetary policy in stabilising economic fluctuations.

And the proposed new fiscal rules:

* In normal times, once a headline surplus has been achieved, the Treasury's mandate for fiscal policy is a target for a surplus on public sector net borrowing in each subsequent year.

* For the period outside normal times from 2015-16, the Treasury's mandate for fiscal policy is a target for a surplus on public sector net borrowing by the end of 2019-20.

* For this period until 2019-20, the Treasury's mandate for fiscal policy is supplemented by a target for public sector net debt as a percentage of GDP to be falling in each year.

These targets apply unless and until the Office for Budget Responsibility (OBR) assesses, as part of its economic and fiscal forecast, that there is a significant negative shock to the UK. A significant negative shock is defined as real GDP growth of less than 1 per cent on a rolling 4-quarter-on-4-quarter basis. If the OBR assesses that a significant negative shock:

* occurred in the most recent 4-quarter period;

* is occurring at the time the assessment is being made; or

* will occur during the forecast period, then the rules are suspended until the budget returns to surplus.

In other words, the government's target is to achieve a 'headline' surplus (negative Public Sector Net Borrowing) by 2019-20; and a surplus in every year thereafter, unless GDP growth falls below 1 per cent, either retrospectively or at any time during the forecast period.

Implications for the surplus target in 2019-20

Let us look first at the implications for the period to 2019-20. As set out above, the projected path of public spending involves very demanding spending reductions, but these appear considerably more realistic than those set out in the March Budget. Taken together with forecast tax revenues, they are consistent with achieving a headline surplus by 2019-20, and in that sense consistent with the new fiscal rule. But there is of course considerable uncertainty, and the point of the fiscal rules is to provide a good guide to (and constraint on) policy under uncertainty. In this light, the new rules are considerably more constraining than the previous version. The surplus target will remain, until and unless GDP growth (actual or forecast) falls below 1 per cent.

[FIGURE 2 OMITTED]

What might this mean in practice? Suppose that growth ran at a steady 1.5 per cent annually for the next four years, while never dipping below the 1 per cent mark. This is considerably slower than predicted by the OBR, or NIESR's, central forecast, but would be well within the margin of error of past forecasts, as figure 2 shows.

The OBR illustrates one specific scenario along these lines, which it describes as a "history repeats" scenario, and shows indeed that such a scenario would result in a significant missing of the surplus target but without triggering the 'get-out' clause (OBR, 2015b). In this case, the surplus target would still apply. Yet GDP at the end of the period would be about 3.5 per cent lower than in the central forecast: tax revenues would be correspondingly lower. The fiscal balance would deteriorate by perhaps 30 billion [pounds sterling], depending on the elasticity of revenues to GDP and the operation of the automatic stabilisers. Without policy action, as the OBR points out, "the Government would miss all its current and proposed fiscal targets": the headline deficit would be about 20 billion [pounds sterling] in 2019-20.

The implication, under the new fiscal rules, is clear. In these circumstances, the government would still be bound by the surplus rule: the shortfall in revenues would have to be made up by tax increases or additional spending cuts. This is simply not remotely plausible. It is inconceivable that the government would respond to a prolonged period of slow growth by implementing an extra 20 billion [pounds sterling] in spending cuts (on top of an already very demanding set of reductions); or that (particularly given its commitment not to raise the rates of the main taxes) it would put up taxes by this amount.

In such circumstances, there is absolutely no doubt that (as in the last Parliament, but against the backdrop of a much more elastic fiscal rule) it would, entirely sensibly, allow the progress of deficit reduction to slip again. In other words, for the period of this Parliament, the new fiscal rule is simply incredible, in the strict sense of the term: nobody should believe it. Note that this does not mean that a surplus cannot or will not be achieved under some circumstances, including the central forecast. It just means that it is entirely foreseeable, ex ante, that under a perfectly plausible set of scenarios the government will choose to ignore the rule.

The fiscal rule in the medium term

Suppose however that, as in the OBR and NIESR's central forecast, a surplus is indeed delivered by 201920, and we are now in the 'steady-state' envisaged by the rules. Are they sensible, and how do they match up against the Treasury's objectives for fiscal policy, as set out above? There are three key elements to the target:

* The choice of target itself--a headline, or overall, surplus.

* The fact that it applies to every year, including the current year.

* The 'get-out' clause--if growth falls below, or is projected to fall below, a certain level then the target is suspended.

In Portes and Wren-Lewis (2015), we discuss the various parameters of an appropriate fiscal target. We show that if, as the Treasury makes explicit above, the objective of policy is long-run fiscal sustainability, then a deficit target should be largely based on a simple debt feedback rule, over a long period. That is, policy should aim at converging debt, rather slowly, to a long-run target; although theory provides little guide as to the level of the long-run debt target. (1) However, we also emphasise that political economy considerations are also relevant: the precise choice of a deficit target and whether it should be cyclically adjusted depends on the extent to which a government is subject to 'deficit bias'.

For a government like the UK, where history suggests that there is a moderate but non-zero tendency towards deficit bias, we conclude that "an appropriate framework for the UK, once interest rates rise above the zero lower bound, would be a rolling medium-term target, for the actual rather than cyclically-adjusted primary deficit." We also note that it would be vital to provide for the suspension of the target, and an alternative approach to fiscal and monetary policy, if interest rates hit the zero lower bound. How does this contrast with the actual target?

First, the target is not cyclically adjusted, in contrast to the previous targets, and consistent with our recommendations. This does reduce the scope for manipulation, and puts less weight on the--inevitably uncertain--estimates of the cyclical position; in that sense it is welcome in principle. However, in conjunction with the short-term nature of the target, it may exacerbate the risks described below.

Second, it is for an absolute surplus rather than being expressed as a target for the primary balance (that is, the balance before interest payments). Both in the theoretical literature, and in practical approaches that focus on sustainability, targets are usually expressed in terms of the primary surpluses (for example, in IMF debt sustainability analyses and programmes). This is probably a relatively minor point for the foreseeable future: obviously it is easy enough arithmetically to relate any target for the absolute balance to one for the primary balance.

Third, it is for every year, not a medium-term target. This is perhaps the most important feature of the new target, and the most risky one. The obvious disadvantage is that it reduces the government's ability to respond to shocks--indeed, even beyond that, it potentially obliges the government to respond to shocks by changing fiscal policy in a way which could be perverse, for example by tightening policy in the face of an adverse shock. Theory suggests that the deficit, rather than changes to taxes or spending, should be the 'shock absorber' for temporary shocks: the rule as proposed risks requiring government to respond in precisely the wrong way.

The 'get-out' clause, suspending the target if growth falls below a certain level, is obviously an attempt to respond to this critique; it is analogous to our recommendation that the targets should be suspended if interest rates are at the zero lower bound. Such a provision is undoubtedly required, and the proposed one is certainly better than nothing. However, it has a number of disadvantages with respect to our proposal:

* Our proposal is grounded directly in the theoretical proposition that underlies the UK and most other developed countries' general approach to macroeconomic policy over the past two decades: that in 'normal times'--that is, when monetary policy is functioning normally--it should be the main policy tool to deal with short-term macroeconomic stabilisation. If interest rates are at the zero lower bound, monetary policy is prima facie not functioning normally (and vice versa). By contrast, GDP growth could fall below 1 per cent (or be forecast so to do) for a number of reasons, some of which would warrant a fiscal policy response and some which wouldn't (and equally there are circumstances in which fiscal policy change might be optimal even if GDP growth was not below 1 per cent).

* Indeed, this tension, and indeed contradiction, is illustrated by the government's own targets for this Parliament. As set out above, the Charter explicitly states that the period 2015-16 to 2019-20 does not yet represent 'normal times', and, implicitly, that this means that an immediate surplus is not yet required. Yet GDP growth is forecast to comfortably exceed 1 per cent throughout the period. So it is clear that this indicator does not, in practice, currently provide a correct signal of whether we are in 'normal times' and therefore need a surplus; so why should it always do so in the future? (2)

This combination of a target set on a continuous (year-by-year) basis and a poorly designed get-out clause appears to be at best risky; this risk is accentuated by practical issues relating to measurement and data revisions.

In addition, the fact that the target applies in every financial year substantially increases the incentive for the government to engage in short-term manipulation of the fiscal aggregates, by artificially shifting expenditure or revenue between years. This is a real concern; even under the old (rolling target) regime, the government did exactly this in the 2013 Budget (the 'cheque is in the post' strategy). (3) Previous governments also engaged in similar manipulation. Perhaps more serious than this short-term game playing are changes which improve the fiscal position in the short term at the expense of long-term fiscal sustainability. As the OBR noted in the March 2014 and March 2015 Economic and Fiscal Outlook, this was the case for a number of the major changes to the tax treatment of pensions in the last Parliament. Of course, any numerical target will, at the margin, distort incentives, but the risks here seem to be substantially higher than for a target of the form we proposed.

What about the level, as opposed to the principle and structure, of the target? Leaving aside the short-term transition to the surplus, theory has rather less to say about whether an absolute surplus is the appropriate medium-to-long-term target. It is of course clear that in itself a headline surplus (resulting in the reduction of debt in cash terms) has no economic significance or rationale whatsoever, so it is clearly not 'optimal' in any sense. But is it any better or worse than a 1 per cent deficit, or a 2 per cent surplus? That depends on what we think is a sensible speed of adjustment to a longterm debt target level.

The Treasury's justification of the target of an absolute surplus was set out in the 2014 Budget (HM Treasury, 2014). This argued that, given the likelihood that the UK would, in the foreseeable future, again experience an adverse economic shock, an ambitious target was needed to ensure that the path of debt reduction was resilient against such a shock. However, this analysis made no attempt to evaluate the costs and benefits of more or less rapid debt reduction, or to justify the particular choice of target. Even less convincingly, in its latest publication on this topic (HM Treasury, 2015b) the Treasury claimed that "Running a surplus on the headline measure of borrowing is the only sustainable way to bring down debt as a share of GDP in the long term." As a matter of arithmetic, of course, this is simply wrong.

In fairness to the Treasury, there is no definitive answer here. But, consistent with our reading of the theoretical literature, recent IMF research (Ostry et al., 2015) suggests that the costs of aggressive debt reduction (in terms of reduced growth and investment now) are rather high, while the 'insurance' benefits of a low debt level (in case of a future crisis) are rather low. Moreover, there is no well-defined methodology for setting a long-run debt target for the majority of countries which, like the UK, have considerable 'fiscal space'. The implication is that debt reduction, while worthwhile, should be very much a subordinate objective compared with sensible macroeconomic and fiscal policy management in the short-to-medium term. Perhaps the most that can be said is that the case for a surplus anytime soon has certainly not been made, and the weight of the existing evidence and analysis suggests that it is not likely to be optimal, but considerably more analysis with direct applicability to the UK is required: this should be a priority for the Treasury and OBR (see Wren-Lewis, 2015).

Finally, one major criticism of the choice of the headline--rather than the current--budget balance as the target is that it may provide a disincentive to public investment, since it draws no distinction between the two. Since, for obvious reasons, the case for funding public investment via borrowing is much stronger than for current spending, while in practice public investment is rather easier to cut in the short term, this may provide precisely the wrong incentives. This would appear to run directly counter to the Treasury's objective of the "promotion of intergenerational fairness"; something that is not addressed anywhere in the new mandate, and which a number of commentators have pointed out has been notably absent from recent policy decisions. As Portes and Wren-Lewis (2015) note: "there are nevertheless clear differences between government consumption and investment in terms of intergenerational equity".

We suggest dealing with this not by incorporating it in the headline target but by a separate target for public investment, and the government does have an objective along these lines, but without anything like the same prominence, and currently set at a low level by historical and international standards. This is particularly perverse when, as now, long-term interest rates are very low by historical standards. The risk of the target continuing to aggravate the UK's long-running failure on this score appears high; it would be sensible to increase the investment target substantially, and to relax the surplus constraint to accommodate this. This issue is further discussed by Smith (2015), who concludes that "the UK government ... should not now saddle itself with a rule that codifies the error."

Expanding the role of the OBR

Assuming that the government intends to proceed with a fiscal mandate broadly on the lines set out in the Charter, the risks identified above emphasise the importance of a further set of recommendations set out in our paper, and absent from the Charter, related to the expansion of the role of the OBR. This could help, in particular, to ensure that long-run issues relating to fiscal sustainability and intergenerational equity do not suffer at the expense of short-term targets, as well as minimise the incentives and opportunity for the government to engage in short-term manipulation of the fiscal aggregates. In particular, I would argue that there is a strong case for the OBR to be given specific remits to:

* assess, at least annually, the long-run fiscal sustainability of the government's tax and spending plans over the medium-to-long term. At the moment, the OBR does produce an annual Fiscal Sustainability Report, but it relies on its own stylised assumptions for the path of spending after the forecast period. In order to do this in a meaningful way, the government would be required to provide the OBR with the basic assumptions needed to do this, for example the long-term strategy towards the financing of the NHS. In particular, the OBR should be asked specifically to identify whether any policy changes have improved the probability of meeting the short-run surplus target at the expense of long-run sustainability.

* As an adjunct to this, the OBR could be asked to assess the implications for intergenerational fairness and intergenerational equity of the government's tax and spending plans, either based on some form of generational accounting approach, or on analysis of specific sets of policies or policy changes (for example, the pension system or the student finance system).

* The OBR should be required to certify, at each fiscal event, that government is observing the spirit as well as letter of the fiscal rules, in particular that it is not artificially switching revenue or expenditures between years.

The Treasury has just announced a review of the role of the OBR (HM Treasury, 2015c): it is to be hoped that these issues are properly addressed.

NOTES

(1) The controversy over the contested findings of Reinhart and Rogoff (2010) also suggests that empirical evidence does not point to a specific optimal target or threshold for public debt; any sensible target will be country and context-specific.

(2) Presumably the government would argue--correctly--that we already have a high deficit, and excessively rapid adjustment would impose too high a cost. But there is no reason to believe such circumstances might not recur without necessarily triggering the get-out clause.

(3) http://www.bbc.co.uk/news/business-21867026.

REFERENCES

HM Treasury (2014), Budget 2014, Annex B, https://www.gov. uk/government/uploads/system/uploads/attachment_data/ file/293759/37630 Budget_2014_Web_Accessible.pdf.

--(2015a), Charter for Budget Responsibility, HM Treasury, July.

--(2015b), 'A country that lives within its means', Spending Review 2015, https://www.gov.uk/government/uploads/system/uploads/ attachment_data/file/447101 /a_country_thatJives_within_its_means.pdf.

--(2015c), Written statement made by The Chief Secretary to the Treasury (Greg Hands), 11 June, HCWS3I: http://www. parliament.uk/documents/commons-vote-office/June%20 2015/11%20June/3-Chancellor-Fiscal.pdf.

Institute for Fiscal Studies (2015), 'Summer post-budget briefing', July, http://www.ifs.org.uk/uploads/publications/budgets/ Budgets%202015/Summer/opening_remarks.pdf.

Office For Budget Responsibility (2015a), Economic and Fiscal Outlook, March: http://budgetresponsibility.org.uk/economicfiscal-outlook-march-2015/.

--(2015b), Economic and Fiscal Outlook, July, http://budgetresponsibility. org.uk/pubs/July-2015-EFO-234224.pdf.

Osborne, G. (2015), 'Speech on the economy', September: http:// blogs.spectator.co.uk/coffeehouse/2013/09/george-osbornesspeech-on-the-economy-full-text/.

Ostry, J.D., Ghosh, A.R. and Espinoza, R. (2015), 'When should public debt be reduced?', IMF Staff Discussion Note, June: https:// www.imf.org/external/pubs/ft/sdn/2015/sdn 1510.pdf.

Portes, J. (2014), 'Fiscal policy and the recovery', April: http:// niesr.ac.uk/blog/fiscal-policy-plan-and-recovery-explainingeconomics#.VaJEd_kl-Pt.

Portes, J. and Wren-Lewis, S. (2015), 'Issues in the design of fiscal policy rules, The Manchester School (forthcoming), previous version at: http://niesr.ac.uk/sites/default/files/publications/ dp429.pdf.

Reinhart, C.M. and Rogoff, K.S. (2010), 'Growth in a time of debt', American Economic Review Papers and Proceedings, 100, pp. 573-8.

Smith, A. (2015), 'George Osborne is committing to the wrong target', July: http://niesr.ac.uk/blog/george-osborne-committing-wrong-target#.VauA6_kl-Ps.

Wren-Lewis, S. (2015), 'Should we aim for budget surpluses?', June: http://mainlymacro.blogspot.co.uk/2015/06/should-we-aim-forbudget-surpluses.html.

Jonathan Portes, National Institute of Economic and Social Research; Centre for Macroeconomics: and Senior Fellow, ESRC UK in a Changing Europe Programme. Email: j.portes@niesr.ac.uk. This paper owes much to discussions with Simon Wren-Lewis (University of Oxford) and reflects arguments further developed in Portes and Wren-Lewis (2015, forthcoming). I am also grateful to Angus Armstrong, Alex Bryson and Simon Kirby for comments.