Heikensten: Swedish economic policy in the EMU
I should like to thank the Academy of Engineering Sciences for inviting me to talk about Swedish economic policy in the event of full participation in the European Monetary Union. I am, of course, particularly pleased to do so as a newly-elected member.
Those of you who have visited the Continent since the turn of the year will no doubt have come into physical contact with the euro. Having previously existed only as an electronic means of payment in financial markets, the euro is now also a reality in the purses and wallets of the citizens of twelve of the European Union's fifteen member countries.
In Sweden this has revived the discussion about EMU, which makes it particularly stimulating to have an opportunity to talk about some related issues. Over the years in my present job I have made ten or so EMU speeches, generally in the light of what I have learnt and experienced by working on economic policy in Sweden as well as in the framework of EU cooperation. Interest in the subject has certainly varied but conditions now seem to be good again for a broad discussion about Sweden and the monetary union. This is a discussion that everyone has reason to encourage before the time comes to make the political decisions.
For clarity's sake I should perhaps add that I am speaking in a personal capacity, not expressing an official position on behalf of the Riksbank.
Some new conditions for economic policy
If Sweden were to adopt the euro, responsibility for monetary policy would be transferred to the European Central Bank (ECB), which acts, not in the interest of particular countries but for what is appropriate for the euro area as a whole. Thus, Swedish economic policy would no longer have recourse to the short-term interest rate and the exchange rate. Tendencies to overheating and inflation just in Sweden could no longer be countered with the monetary alternative of raising the interest rate here, just as the interest rate could not be lowered to mitigate a marked slowdown in our economy. This is clearly a change in the conditions for economic policy that needs to be analysed and discussed in advance of a decision to adopt the euro.
Having said that, the issue should not be seen too dramatically. The set-up for economic policy in a monetary union is much the same as in a country with a fixed exchange rate. In Sweden that was actually the case in 73 of the past 100 years, with results that were excellent in some periods and considerably worse in others. It is important to learn from those experiences. What distinguishes a monetary union is that the exchange rate is locked irrevocably. That means there is no room for the type of misguided economic policy that characterised the 1970s and '80s, for example, when Sweden used repeated devaluations in an attempt to rectify the situation. Such failures must be avoided at any cost.
Against this background it is not remarkable that the discussion about economic policy with EMU has focused on stabilisation policy. Even so, it is important to start from a broader perspective. The issues to do with stabilisation policy have featured prominently in the Swedish debate. This has to do with a strong Keynesian tradition and the ability of Lars Calmfors and his former colleagues to place the so-called insurance argument in the centre of the whole discussion about a monetary union. The latter naturally has to do with the fact that the very dramatic crisis which hit Sweden in the early 1990s was so closely connected with stabilisation policy. It is hardly surprising that in many other countries the discussion took a very different course before they adhered to the EMU.
It is extremely important that our discussion about economic policy - with or without the euro - does not overlook or play down the supply side of the economy. Countries that create conditions for a smooth adjustment of labour markets and a favourable development of productivity will almost certainly do well as members of the monetary union. Measures to this end are naturally warranted in any event - they promote prosperity. But they also alter the conditions for stabilisation policy and reduce the need for 'insurance'.
In this context it may be worth recalling that the overheating that led up to the crisis in the early 1990s was by no means primarily due to the growth of demand being notably strong, though demand was certainly too high in relation to the economy's productive capacity. Annual GDP growth from 1985 to 1990 was actually no more than 2.4 per cent. What really mattered was poor conditions for growth; productivity rose only 1.1 per cent and employment in hours only 1.3 per cent. If productivity growth in the late 1980s and early 1990s had been half or one percentage point higher, things could have been very different.
Our tendency to concentrate the EMU discussion on stabilisation policy means that a number of important points about EMU cooperation are overlooked. They concern the longer run and aim precisely at improving conditions for growth and employment in this horizon. EMU has an important role here. Trade will be facilitated along with other cross-border activities such as investment. Simplified price comparisons can led to stronger competition. We will also have a broader and deeper capital market. Moreover, EMU implies better conditions for a stable development in the Union as a whole. Good examples are the crises in Asia and Russia around 1998 and last year's abrupt slowdown in the American economy; had it not been for EMU, Europe's financial markets would have been more turbulent.
Policy frames in the European Union
Before discussing economic policy in Sweden in the event of a transition to the euro, something should be said about the European frames which this policy would have to comply with. These frames already affect the conditions for Sweden's policy and will presumably do so even more in the future.
Of the economic policy processes in the European Union, the most prominent are contingent on the Stability and Growth Pact from 1997. The primary purpose of the Pact is to ensure that budget policy in the member states is conducted so that no country encounters serious problems with debt because that could lead to costs for the Union as a whole and, in the worst case, to other countries being forced to bail out the 'sinner'. The Pact contains binding rules and in well-defined extreme cases could even result in sanctions. In addition, there is a number of processes or rules whereby peer pressure and benchmarking are used to speed up the EU's structural reforms. The central policy instrument is the general economic policy guidelines, which can be likened to a budget statement. Other instruments are, for example, the Luxembourg, Cardiff and Lisbon processes. All the above applies to Sweden, whether or not we adopt the euro, though the terms of the Growth and Stability Pact do differ in this respect; the rules for sanctions, for instance, do not apply as long as we are outside the euro area.
There is much to be said both about the actual construction of the processes and what they have achieved to date. Moreover, knowledge of these matters is generally lacking in Sweden. Today, however, I shall be very brief. As I see it, there is a tendency towards an increased coordination of fiscal and structural policies. This takes many forms. The issues are discussed together, both at the highest level and further down among civil servants and experts. It is partly a learning process, an exchange of thoughts about good examples, a communication of ideas and so on. New, more specific knowledge is also being generated. At present, for instance, common principles are being drawn up for the measurement of budget outcomes and capacity utilisation. This is necessary for effective surveillance of compliance with the Stability and Growth Pact. Similar work has begun in the field of pensions. These quotidian efforts should not be underrated. They help to make many problems more understandable and promote a consensus about how they should be handled.
Sweden has much to gain from this kind of international collaboration. To exemplify how international discussions can influence matters I can mention that when we at the Ministry of Finance were aiming to alter the economic policy agenda in the late 1980s, we were greatly inspired by the work that had been done under the auspices of the OECD and others. In my opinion, the policy realignment owed at least as much to our international contacts as to the successful opinion-forming work by, for example, the Centre for Business & Policy Studies (SNS).
Many of the problems and measures that are being discussed are of a national nature; they vary between countries and it is in each country's interest to solve them. Moreover, due to differences in the initial situation, the solutions may have to differ, too. One should therefore be cautious about imposing universal mandatory rules. Furthermore, solutions will presumably become more uniform as a result of increased trade and competition. It is likely, however, that fiscal and budget polices will be further coordinated, though this process will not necessarily be driven solely by principles and intellectual arguments. I am more inclined to guess that circumstances will bring this about by degrees as new problems, stemming from the member states sitting in the same boat, have to be tackled, perhaps in some cases at rather short notice.
Fiscal policy in Sweden
The problems for stabilisation policy
So what about the problems for stabilisation policy that arise if the euro is adopted? We would clearly have to do without an independent monetary policy. By itself, this may imply a risk of larger cyclical fluctuations and it would also make a proper crisis more difficult to manage with measures of economic policy. As was the case in the early 1990s, a crisis can arise mainly from an inappropriate domestic policy; it can also result from an external shock that hits Sweden more strongly than the average euro country; a third possibility is the failure of one or more Swedish companies of vital importance for our economy.
There are other ways in which the conditions for economic policy can be affected. As a euro participant, a country has somewhat more room for manoeuvre in fiscal policy. This is really only the other side of the arguments behind the Stability and Growth Pact. Financial markets would have less reason to follow developments in Sweden, with the attendant risk that fiscal and budget policies become less sustainable. Another question concerns the effect on wage agreements. Views differ on this. Some count on a more disciplined approach if Sweden adopts the euro; otherwise a failure could then be very costly, with years of unduly weak competitiveness and growth. Others argue that as things are, the connection between wage outcomes and the interest rate in Sweden is more direct. They sometimes refer, for example, to the outcomes of the latest round of wage negotiations and the effect on them of interest rate policy in autumn 1997.
This undoubtedly warrants a closer look at what the framework for stabilisation policy should consist of. A major problem in this context is clearly that, for a number of familiar reasons, it is more difficult to use fiscal than monetary policy for active economic stabilisation. A crucial problem in both cases is not knowing in which direction the economy is moving. Here it will suffice to recall what most of us here were saying about the economy two years ago. Another, related problem has to do with the time lag before measures have an impact. In the case of monetary policy we usually allow for a lag of one to two years. Arriving at decisions also takes time, particularly in fiscal policy. Moreover, decisions are often swayed by other considerations than stabilisation policy, which is just why the implementation of monetary policy has been assigned to an independent institution.
Firm principles for budget policy
This speaks for the importance of not exaggerating what can be achieved with fiscal fine-tuning. What matters is that the fiscal policy framework is such that adequate reserves are built up. The combination of a stable public sector financial surplus over the business cycle and low debt provides a margin for the government finances to deteriorate in the normal way when activity is weakening and let the automatic stabilisers work. If that functions properly, we will have made good progress.
Given adequate reserves, Sweden will be in an unusually good position in this respect because our extensive public sector, high replacement levels in transfer systems and high taxes mean that the budget balance fluctuates more than in other countries. At the same time, it should perhaps be mentioned that an economy contains many other types of stabiliser, not all of which are necessarily public. A system like ours, with high income taxes and transfers, naturally also influences the extent to which private safety-nets or reserves are built up. This includes everything from saving by households to contractual unemployment insurance.
For the automatic stabilisers to function in full during future downturns it is important to stick to the current rules for fiscal policy and ensure strict compliance when times are bad as well as good. There may also be grounds for being even more ambitious about the budget surplus and aim for a level of, say, 3 per cent over the business cycle. Debt could then be repaid more quickly, giving us more room for manoeuvre in the future. The spending ceiling has functioned well both by breaking what was previously an almost automatic increase in government expenditure and by maintaining discipline in the central government finances. The same applies to the requirement that the local government finances are balanced; there is less talk of this requirement in contexts like the present but it has been an essential part of the framework that has held back the expansion of public spending in recent years. It is therefore important that these rule systems are not eroded. On top of all this there may be a case for some changes in the arrangement of budget work. Some of the proposals put forward recently by Svante Öberg can be considered as a way of strengthening the process still more. It is important, for example, to ensure that the budgeting margins are in fact handled so they can be used as intended.
Notable progress has unquestionably been made in this field in recent years. This may be a problem in the ongoing discussion, which often seems to assume that consolidation is something that happens almost by itself. I do not believe that is so. The reason why everything has gone so well just in recent years seems to me to be that the crisis in the early 1990s is still in everyone's minds and therefore makes a firm policy easier. Moreover, growth as well as inflation have been unexpectedly favourable for three to four years and this has consolidated budgets much sooner than had been foreseen.
Looking ahead, the challenges are considerable. If the quality of publicly financed services such as school education and the health service is to be maintained at an acceptable level in the longer run, this will be rather costly. That is because productivity in activities of this type, be they produced publicly or privately, will not rise as quickly as in the production of goods. Moreover, Sweden has an ageing population. Together with any difficulties in sticking to a tax level that is internationally high, this would severely test the public finances. Under these circumstances it will be exceedingly important to make a thorough scrutiny of the social security system's design; this is a question that has been on something of a back burner, what with the crisis for stabilisation policy in the 1990s and the achievement of a more lasting solution to the issue of pensions. In such a discussion about social security systems that are effective and sustainable in the long term there is a place for the issue of cyclical stabilisation. Systems that are fully or partly funded, publicly or privately, can function well as stabilisers. Thus, it is mainly in this context that I would like to place the discussion that has taken place about various types of buffer fund. But the need to reform the social security system is primarily connected, not with stabilisation policy but with being able to guarantee a high quality of publicly financed services without raising the tax take.
Measures for influencing the business cycle
Finally there are the questions of whether a more active, discretionary fiscal policy would be advisable and whether, to function better in the future, it may need to be supported by new institutional arrangements. First, let me stress that, for reasons I mentioned earlier, I do not have much faith in fine tuning. Of course that does not rule out trying not to accentuate fluctuations as far as possible when measures are taken primarily for other reasons. The central issue, however, is how more pronounced problems or crises for stabilisation policy are to be handled.
One approach to this is to relate once again to our own experiences. As I see it, we have had to face problems of two types. One concerns a lack of knowledge about the direction in which the economy is moving. A dramatic example of this is the crisis in 1990, when neither the problem that was looming with the real interest rate nor the crisis in the financial system was identified in time by virtually any Swedish observer. The other problem is political and has to do with the difficulties in keeping policy sufficiently tight over the business cycle, particularly when demand exceeds long-term supply. A good illustration of this problem is the late 1980s.
Perhaps one should mention a third problem, namely that automatic stabilisers affect demand regardless of the nature of the preceding shock. If demand in general has weakened on account of, say, a share price fall that has affected household wealth, all will be well; a part of the loss of demand will be compensated automatically. Things will be different if the shock comes from the supply side, for instance as an oil price rise or a change in productivity. There could then be grounds for a discretionary policy move to adjust demand to the change in production capacity; this could be more difficult in the absence of an independent monetary policy. As a euro participant, however, Sweden would be in the same boat as other member states as regards many of what appear to be likely supply shocks, for instance oil price fluctuations.
As regards the lack of knowledge, there are clearly no simple institutional solutions. But this does not mean that the institutional frameworks for conducting policy do not affect the accumulation of knowledge. As I mentioned earlier, the Stability and Growth Pact has led to a rapid development of EU work to arrive at a better understanding of the interaction of public budgets and capacity utilisation. In Sweden, the acquisition of knowledge about potential growth, the output gap and so on has been stimulated in that the Riksbank has been made accountable for safeguarding the value of money and we have placed this mandate in a clear intellectual framework that calls for deeper analyses. This in turn has had consequences for how both the Ministry of Finance and the National Institute of Economic Research are now working on these matters. Another reason why Swedish authorities have reason to be more involved in a broad discussion and analysis of questions such as these is perhaps that euro participation could lead to the financial markets becoming less interested in analysing specifically Swedish conditions.
I consider that the importance of a high level of competence in such institutions as the Finance Ministry, the Riksbank and the National Institute should be emphasised. Good links between these institutions and the academic world are also desirable; in a European context I consider that in this respect we are already in the forefront. It is likewise important that we participate in the international work on these issues, primarily in the context of the European Union and the ECB but also in the OECD and the IMF. And although we are doing relatively well internationally, a good deal is being accomplished elsewhere that we can try out here in Sweden. As a member of the Governing Council of the ECB, the Riksbank will have cause to go on working on these issues at the European level. I also think it would be natural to continue to publish the Inflation Report twice a year; that would be a way of presenting analyses of the interplay between cyclical activity, monetary policy and the government finances and thereby contribute to the discussion of economic policy. A qualified ongoing analysis is needed for a small country like Sweden to play a meaningful role in the international cooperation. Besides the traditional issues of stabilisation policy, there are grounds for the Riksbank continuing to be deeply engaged with developments in the financial sector, for instance with the Stability Report. A conceivable evolution of the existing structure as regards the basis for fiscal policy could be a more direct involvement of the Economic Council that is attached to development work in the Ministry of Finance.
Handling the political problems to do with a consistent, long-term approach to stabilisation policy is considerably more complicated. In practice it seems to be mainly a matter of ensuring that budget policy sticks to the rules in upward phases, both in order to secure reserves for down-turns and to avoid overheating, with its negative effects on wage formation, costs and so on. One possibility that has featured in the public discussion would be to extend the function of the Economic Council, for instance by formulating opinions about the formation of policy on behalf of the Riksdag (Sweden's parliament). At a general level I can see no crucial objection to this; one more voice in the public discussion can do no harm. But as soon as one goes further than that, difficult issues arise.
Time does not permit a long discussion of this question, which is not all that central. I just want to say that it touches on aspects of democracy that are likely to become even more tangible in that, if Sweden adopts the euro, monetary policy will be conducted well beyond our borders. The difficulties are compounded in that in questions such as these - taxes, cyclical activity, the direction of fiscal policy - the distinction between knowledge and values tends to be indistinct. To exemplify this I can cite a discussion I had two years ago with representatives of the Ministry of Finance. At that time they considered there was a case for cutting taxes and argued on the basis of estimated resource utilisation that this was feasible without risking capacity restrictions. Our estimations suggested otherwise, though the differences should not be exaggerated. We still do not know who was right. But what we do know is that economic activity has declined far more than any of us foresaw and this has been accompanied by higher inflation.
Concluding remarks
Now for a brief summary of what I have said.
Economic policy as a euro participant is not all that different from economic policy today. When discussing stabilisation policy it should borne in mind that the best way for us to achieve a favourable development - inside or outside the euro area - is to create conditions for a smooth adjustment to various markets, primarily the labour market, and for a good productivity trend. In these respects there are measures we ought to take in any event. But euro participation can help us to move in the right direction.
By adopting the euro we would relinquish our national monetary policy. That would make the formation of fiscal policy more crucial. There are risks here, as we know from our own experience in the 1970s and '80s, when policy was neither consistent nor focused on the long term. As a result, cyclical fluctuations were accentuated and the government finances gradually weakened. That limited the room for action when Sweden entered the 1990s.
The main thing to do to prevent new crises for stabilisation policy is to develop and possibly reinforce the framework for budget policy that has applied in recent years. Government debt can then be run down more quickly and reserves can be built up so that the automatic stabilisers can be left to do their work. An important issue in this context is the construction of the social security system. Changes will be needed and there may well be reasons for increasing the element of funding.
It may be asked whether some institutional changes are also needed to facilitate a discretionary fiscal policy. This is certainly not a minor question but it has taken up too much of the discussion about economic policy. The possibility of implementing a discretionary fiscal policy in a meaningful way is limited. The discussion should therefore concentrate on what can be done to manage situations when more serious shocks are likely to occur.
The problem here has been a lack of knowledge, plus the difficulties in mobilising sufficient support for a restrictive policy when the economy is overheating. There are presumably some things that can be done to improve our knowledge, though it is mostly a matter of trimming an existing structure. But that must be done in the light of the limited competence for empirical economic analysis in Sweden and the numerous challenges that lie ahead in other respects than stabilisation policy.
Perhaps a sharper analysis and discussion could also be a help with the political problems. In this context, an economic council along the lines put forward in the public debate might possibly play a part.
