Persson: Monetary policy and competition
Thank you for the invitation to come here to Roslagen Sparbank's annual general meeting.(1) Today, I do not intend to discuss current monetary policy, but rather to concentrate on an area that is of great significance for inflation and monetary policy, namely the competitive conditions in the Swedish economy.
The objectives for monetary policy are clearly stated in the Sveriges Riksbank Act - to maintain price stability and to promote a safe and efficient payment system. The Riksbank has further defined the first of these objectives in an inflation target, i.e. that the annual rate of change in consumer prices shall be 2 per cent, with a tolerated deviation interval of ± 1 percentage point. Interest rate decisions influence price trends partly via the effects on resource utilisation, which means that the impact of the decisions has a certain time lag. It is partly for this reason that monetary policy is based on the total assessment of how the rate of price increase will develop one to two years ahead.
There are a number of factors in the economy that affect inflation, but over which the Riksbank has no control. These include the exchange rate, fiscal policy, international economic developments, potential growth, the functioning of the labour market and competition conditions on the product markets. Today I intend to primarily focus on the relationship between monetary policy and competitiveness.
Let me first of all state that the Riksbank has no responsibility for competition issues - our responsibility is monetary policy - and our most important instrument, the repo rate, does not affect competitive conditions on the different markets. However, changes in competitive conditions can have significance for price trends on certain sub-markets.
From textbook to reality
If there were perfect competition and total information, conditions that only prevail in university textbooks, market prices would be determined by companies' marginal costs. As there is competition between a large number of producers on every market, no company has any power over pricing and all consumers meet the same prices. On a market characterised by perfect competition, more companies would establish themselves as long as the prices exceeded the marginal costs and they could make a profit. As there would be no entry barriers in this simplified, but pedagogical textbook world, and no lack of skilled entrepreneurs, the profits of all companies would be zero when the market was in balance. In a market with perfect competition, the impact of a cost change (possibly resulting from a weaker exchange rate or a rise in the rate of wage increase) would be total, as the price is set as equal to the marginal cost.
Reality is, of course, rather different. Many markets have only a small number of companies competing, partly because production requires large investments or because there are other barriers to establishment on the market. Nor are there unlimited entrepreneurial skills. It is costly for consumers to obtain information, which means that no one has total information. On this type of market, prices deviate from marginal costs and not all consumers meet the same price levels. Similar products can fetch different prices because they have different brand names or are sold in different regions. These companies will normally make a profit. Companies that operate on a market with imperfect competition also tend to be less cost-conscious.
In a market with imperfect competition, price changes involve a more long drawn out process. The effect of changes in marginal costs also depends on factors such as demand and the initial profit margin. Most empirical studies indicate that tougher competition (measured as the degree of company concentration) leads to a more rapid price adjustment after a shock, but there are also studies that have reached the opposite conclusion.(2) In a market with imperfect competition, expectations play an important role. If companies believe that a change in marginal costs is only temporary, they will be less likely to allow it to affect their prices, while more permanent changes will have a quicker impact. The low impact from changes in the exchange rate in recent years is merely one example of how changes in marginal costs that are perceived as temporary tend to affect profit margins rather than consumer prices.
It is possible to observe how competition affects inflation through a concrete example. Let us take a look at the book trade. If we assume that the Internet means that more consumers will choose to buy cheaper books over the net instead of going to their local bookshop, then bookshops will be forced to lower their prices in order to retain customers. The next stage is for the bookshops to try to rationalise their operations. Productivity in the book trade will increase. If they are successful in their work, they can retain their profit margins. Another possibility is that they will not require such high wage increases in the next round of wage negotiations as they would have done otherwise, or that they will pay less when they purchase books from the publishing companies.
When a market is exposed to competition, this can have both direct and indirect effects on prices and wages. The price-dampening effects of increased competition can contribute to subduing inflation for a period of time. One example of this is the deregulation of the electricity and telecommunications markets that contributed to the low inflation rate during 1999 and 2000, although the price-dampening effects disappeared at the end of 2000. Since then, electricity and telecommunications prices have contributed to higher inflation. Increased competition means that companies have to use their resources more effectively and be more cost-conscious. Growth and productivity will therefore increase. Lasting effects on productivity growth could affect the potential growth rate, i.e. how quickly the economy can grow given the amount of labour and capital available and without a rapid rise in inflation. (3)
However, this type of development is not without problems. If there is a series of improvements in productivity, the result will be higher growth and limited pressure on prices. During such periods there may be slightly exaggerated optimism over long-term sustainable growth in profits and income, which will also tend to affect prices of various assets, such as shares, houses and apartments. If the central bank makes the mistake of conducting an overly mild monetary policy during such a period, perhaps because inflation is below the target level for a period of several years, there is a risk that the underlying price pressure will increase and future inflation will soar.
The relationship between potential growth and competition is not unambiguous. It may be the case that a new invention gives rise to a temporary monopoly - partly due to legislation regarding patents and partly because it involves technological progress. Some economists, and in particular Joseph Schumpeter, have even claimed that the endeavour to achieve these temporary innovation monopolies is one of the most important driving forces behind technological developments (which together with labour supply comprise the motor for long-term potential growth). One example is the pharmaceuticals industry, where the development of new products would probably not be possible at the same rate if new medicines were not protected by patent laws. Incomplete competition is thus important for ensuring that product development takes place within research-intensive operations. There may even be special circumstances that make production most effective under limited competition. This applies in particular to various types of infrastructure, for instance the trunk network for electricity and water, which are what is known as natural monopolies, where it is more effective to society to only have one producer.
These assertions are important to monetary policy. The Riksbank's target is expressed in terms of the rate of change in consumer prices, not in terms of maintaining a particular price level. It is important to distinguish between the effects on the price level and the change in the price level. If competition is incomplete and leads to a low level of cost consciousness or high profit margins, it may contribute to high price levels, but will not necessarily affect the rate of change in prices. It is conceivable that a country could have a high price level to start with, but a low expected inflation rate. An example of this is that several countries with high price levels, such as Sweden, Switzerland and Japan, have been among the countries with the lowest inflation rates in recent years.
The relative price level in Sweden
The price level in Sweden is approximately 28 percentage points higher than the EU average and Sweden is thus, along with the other Nordic countries, one of the most expensive countries in the EU. (4) There is a considerable spread of price levels within the EU, as illustrated in diagram 1. The price level in the most expensive country (Sweden) is twice as high as that in the cheapest country (Portugal). Sweden in general has higher prices on goods and services not traded across borders, while the price level for trade goods such as clothes and furniture is below or in line with the EU average.
The fact that there exist price differences between countries is partly due to structural factors. One important factor is the income level (richer countries usually have higher prices). Taxes, wages, population density and changes in the exchange rate are also of considerable significance. Sweden has, for instance, higher VAT and unit labour costs, i.e. payroll expense including social security contributions adjusted for productivity, than the EU average. Sweden's geographical situation and the fact that we are a sparsely-populated country mean that we are also affected by higher transport costs, lower economies of scale and thereby have higher prices.
It is estimated that approximately one half of the Swedish price level difference from the EU average can be explained by structural factors. The other half is attributed to a lack of competition. (5) Calculated in krona, this means that a family with two children pays approximately SEK 30,000 more a year for their purchases than they would otherwise. One can therefore say that there are considerable welfare losses connected with weak competition.
How does competition function in Sweden?
Competitive conditions have improved during the 1990s and the relative price level has fallen, as can be seen from diagram 2. The factors behind the improved competition situation are increased internationalisation, deregulation, technological developments, EU membership and the more stringent competition laws introduced in 1993. The fall in relative price levels in 1993 is largely due to a weakening of the exchange rate in connection with the floating of the krona in November 1992, when Sweden became a relatively inexpensive country in terms of euro. Sweden has also had a lower inflation rate than other EU countries, which has contributed to keeping down the relative price level.
The part of household consumption that takes place in Swedish markets exposed to competition has increased during the 1990s - from 25 per cent in 1991 to 32 per cent in 1999.(6) However, this level is still below the EU, where 46 per cent of consumption is in markets exposed to competition. A number of different markets have been opened up to competition over the past ten years and Sweden is in many aspects a model for other countries to follow. Examples of markets that have been exposed to competition are the electricity market, post and telecommunications, the capital and foreign exchange market and a large part of the transport sector. The main areas where competitive pressure in Sweden is tangibly lower than in the EU and prices are higher are the food and housing sectors (and the construction sector). Swedish food prices are just over 20 per cent higher than in the rest of the EU, while housing is around 35 per cent more expensive.(7) These areas of expenditure together comprise more than ¼ of the CPI basket and are also responsible for a large part of low income earners' total expenditure.
The absence of competition in the greater part of the public sector, which constitutes approximately 30 per cent of the Swedish economy, entails both a risk of lower efficiency and that a significant part of the economy is not offered alternatives in the form of new business ventures with the growth and flexibility that small companies can offer. During the 20th century there was a development towards a greater element of competition and use of alternative operating forms within the public services sector. In the central government sector, state-owned monopolies such as the post office, the telecommunications administration and the state-owned energy company have been opened up to competition. Examples of remaining state-owned monopolies are Apoteksbolaget, Systembolaget, Swedish Railways, Svensk Bilprovning, Svenska Spel and Statens Pensionsverk. The county council/local authority sector has now chosen to expose to competition many of the businesses it runs. This includes waste transport, property maintenance, local bus routes, care of the elderly, cleaning, primary healthcare and dental treatment. Studies by the Swedish Association of Local Authorities show that increased competition in the county council/local authority sector operations have increased efficiency and led to lower costs.
One indicator that competition is not functioning satisfactorily is that price levels vary even within Sweden. Western Sweden (Halland and Västra Götaland county council) is on the whole cheaper, while the Stockholm region is more expensive than the country as a whole.(8) In terms of money, these differences entail, for instance, that a family with two children in Western Sweden pays SEK 4,300 less a year for food than a similar family in the Stockholm region.
Why is competition lower in Sweden?
Sweden has more markets with a high level of company concentration. Concentration is higher than in the rest of the EU, particularly within the retail trade, while the difference in the producer sector is not as great. The Swedish Competition Authority has estimated that the higher concentration in the retail trade may explain 2-4 percentage points of the difference in prices between Sweden and the EU.
In the everyday commodities market, 3 retain chains (ICA, KF and Axfood) account for 90 per cent of total sales. These chains control both the retail trade and the distribution system, which makes it difficult for new business to establish. Access to shop premises is also a narrow sector, as the right to establish a business is regulated by the local authorities through the planning and construction laws (PBL). (9) Unlike the rest of the EU, Sweden does not have very many low-price retail chains. At the moment, a German low-price retail chain is planning to establish itself on the Swedish market, which could have a dampening effect on prices.
A high concentration of companies also increases the risk of forming cartels. Recently, several cartels, or suspected cartels, have been revealed in the building, aviation and petrol trades. According to a recently published SIFO survey, 80 per cent of small companies in Sweden and 50 per cent of large companies state that they believe cartels arise very often or quite often in Swedish trade and industry. In other words, consumers may be losing considerable sums of money. With regard to the "petrol cartel", amounts in the order of half a billion kronor were discussed. By eliminating market mechanisms, companies also avoid the competitive pressure that could otherwise ensure they make their operations more efficient and develop new products. In the long term, it also leads to a loss of competitiveness from an international perspective, lower growth and fewer jobs.
The government has proposed measures against unlawful co-operation in the bill on combating cartels that was put forward recently. (10) The bill aims to make the war against cartels more efficient by allowing a company that exposes a cartel in future to avoid paying the fine for damage to competition or to have its fine reduced. The government also proposes a stronger confidentiality protection for those who report or provide information on cartels.
The question is whether this tightening of the laws will be sufficient; the incentives for following the competition laws are still fairly weak. It is individuals (company managers, directors, owners) who prepare and make the decisions on limiting competition in contravention of the law, but the current legislation is fairly weak with regard to personal responsibility. In the USA, on the other hand, legislation is more stringent - the responsible decision-makers can be personally sentenced to high fines or to imprisonment. Perhaps we need to have the same principle in Sweden. The government has also said that if the proposed tightening of the competition laws does not prove sufficient, it may consider designating these matters a criminal offence.
The existence of what is known as non-tariff trade barriers creates difficulties for import competition, which could otherwise have subdued prices. One example is the building materials market, where special national regulations and voluntary product standards make import competition difficult. Sometimes the endeavour to satisfy information and protection interests can lead to trade barriers that hamper competition. The requirement that chemical products such as washing powder and cleaning fluids (which are more than 90 per cent more expensive in Sweden than the EU average) must be registered in a special product register containing information on the exact contents also hampers import competition as most other countries do not have such registers. The fact that the Swedish market is small means that it is not economically viable for foreign manufacturers to make the adaptation required by Swedish rules or legislation. The existence of recycling systems for cans and bottles also makes import competition more difficult in the soft drinks market (soft drinks are approximately 40 per cent more expensive in Sweden than the rest of the EU).
The fact that requirements motivated by concern for the environment could impede competition does not, of course, mean that they should be abolished. On the other hand, this is a good argument for co-ordinating the various regulatory systems in Europe.
The bank sector
As I am today speaking to people who work in the bank sector, I thought I would take the opportunity to say a few words on competitive conditions in this sector.
The bank sector was characterised by extensive regulations right up until the mid-1980s, which acted as an impediment to dynamic development in this sector. Few new operators could enter and competitive pressure was fairly low. However, the deregulation that has occurred since then has radically changed this picture. Today, the Swedish bank market is far more dynamic and a large number of both Swedish and foreign financial companies offer products covering everything from unit trusts to mortgages. Although competition could probably be improved even more, developments are moving in the right direction.
From a European perspective, the Swedish banks currently make relatively low charges for their services and have low profit margins on their deposits and lending - these are in any case no higher than in the rest of Europe. (11) Today the Swedish banks are in the forefront with regard to technology and have rationalised their operations, with fewer offices and employees than banks out in Europe. Despite this, profitability is relatively modest, which indicates that the price level is under pressure and competition is functioning.
However, the company concentration is still very high on the bank market. The four major banks together have a market share of 70-90 per cent within most product segments. A number of "niche" banks have become established over the past year, but they still account for a small, albeit growing, share of the market.
The bank market differs from, for instance, the retail trade in that the confidence and personal contact with the bank is perceived as very important. The customers also face complex choices; it is often difficult to assess and compare different services and suppliers. In addition, establishing new banking and insurance relations involves work for the customer. Small companies may find it particularly difficult to change bank, as a long-term relationship between the company and the bank is an important part of the bank's credit assessment. These factors mean that few customers change banks; only 14 per cent of the population have changed one or more banking commitments during the past three-year period.
The fact that bank customers do not readily change bank may mean that there is less price pressure on certain products than would otherwise be the case. However, regulations and tax legislation may also have the effect of subduing competition. One area that is worth particular attention is saving in funds. Although competition appears to be good for new savings, there is a considerable deadlock effect resulting from regulations on capital gains tax. This helps dampen competitive pressure.
One area that is still an object for old-fashioned, national limitations is cross-border payments. Despite electronic systems and the fact that the EU is supposed to comprise a single market, substantial costs are involved in transferring payments between banks in the different countries.
The EU and EMU
EU membership is one of the structural factors that are assumed to have increased competition in recent years. The idea in creating the EU's common market was to abolish the legal and administrative impediments to trade and thereby increase competition between and within the member states of the EU. This, together with better usage of economies of scale, would cause price levels to converge. It is now 9 years since the common market came into force and we can conclude that the price differences between the member states still remain. There are also considerable differences between the USA and the EU with regard to price spread. Almost one-third of all brands has a price spread of between 60 and 100 per cent within the EU. The corresponding figure in the USA is 9 per cent.
One reason for this is that even if the legal trade impediments have been largely removed, there still remain a significant number of non-tariff trade barriers in the form of, for instance, various standards that segment the national markets. Transport costs and differences in consumer preferences are also factors that counteract market integration. A clear sign of market segmentation is that the price differences between the countries are significantly larger than within the countries.
The EU/EEA countries trade almost 90 per cent with one another. Internal trade is therefore approximately as important or more extensive here than it is in the USA. The fact that the European market is much more segmented than the North American market reduces the efficiency within the EU.
Competition could be improved further throughout the whole of the EU. This is one of the areas where efforts will be made to contribute to improving productivity and growth within the EU to make the union one of the world's most competitive economies in 2010. At the moment, there remains much work to be done before this goal, which was established at the Lisbon meeting in 2000, can be achieved. Some small steps forward were taken at the Barcelona meeting last month. For instance, the countries agreed to open up the electricity and gas markets to corporate customers in 2004 and to open up the common air space by 2004.
The single currency, the euro, could lead to stiffer competition as it will be easier for consumers and companies to compare prices of consumer goods and input goods in different countries when they are expressed in the same currency. The EMU could in the long term be expected to increase trade and integration between the member states. The studies that have been made indicate varying effects on trade from EMU membership for Sweden. Berkeley Professor Andrew K. Rose has estimated that Sweden's trade could increase by 30-50 per cent over the space of a few decades, which would lead to an increase in Swedish welfare of more than ten per cent in the long term. Other economists, such as Professor Torsten Persson at Stockholm University, estimate the trade gain at one fifth of this.
An increased openness towards other countries has a positive effect on growth and on incomes. The European integration should therefore, in my opinion, not be limited to the borders of the EU, but be the first stage towards greater global openness. This openness would also contribute to increased competition, but would not replace a domestic competition policy. Both openness and a tightening up of the domestic conditions for competition would also lead to lower prices and increased welfare.
During the 1990s there were a number of structural changes in the Swedish economy, which are together assumed to have increased potential growth. There were extensive reforms in the taxation, social insurance and pension systems. Deregulation, increased internationalisation, EU membership, a credible low inflation policy and stable public finances are other factors which should also have had positive effects.
It was also proven during several years in the 1990s that a high growth rate was compatible with a low inflation rate. However, the strong growth in demand in recent years has gradually reduced the amount of available resources. The Riksbank concluded in its most recent Inflation Report, which was published on 19 March, that a significant part of the high inflation rate over the past year can be an indicator that resource utilisation was slightly higher than was compatible with the Riksbank's inflation target. During 2001, growth was mainly in sectors with a relatively low growth in productivity, such as the private services sector and public authorities.
At the same time, absence due to illness has increased considerably, which may also have contributed to a poor growth in productivity and rising costs for both companies and the public sector. The rate of wage increase has risen, particularly within the services sectors, which appears to have contributed to pushing up the rate of price increase in the Swedish economy. This is an important reason why the Riksbank raised its repo rate by 0.25 percentage points, to 4.0 per cent, in connection with the publication of the Inflation Report just over one month ago. We also signalled on that occasion that if the trends regarding economic activity and inflation propensity in the Swedish economy remained the same, there may be cause for further increases in the interest rate.
Growth in demand now appears to be on the right path both in Sweden and abroad. Swedish fiscal policy is currently conducted in a very expansionary direction, which made it natural for the Riksbank to wish to ease up on the accelerator to ensure the inflation target is achieved. Some people might think that this target means that the Riksbank is forced to give less priority to employment and production in a situation where a recovery is still in its infancy. However, first one must remember that changes in the repo rate have an effect one to two years ahead. Secondly, there is no antagonism between growth and price stability in the long term; it is potential growth that sets the limits for how fast the economy can grow. Stable prices actually create the best conditions for sustainable growth. The Riksbank cannot affect potential growth. Other operators can do this - trade and industry primarily through investments and the government and parliament, which decide over the regulatory system, education and the infrastructure, etc. This is probably the most important contribution to both good growth and stable prices.
(1) Sara Tägtström, of the Monetary Policy Department, has contributed to the writing of this speech.
(2) For an outline, see Asplund, M. & Friberg, R. (1998), "Links between competition and inflation", Sveriges Riksbank Quarterly Review 3, Sveriges Riksbank.
(3) See, for instance, Dutz, M. & Hayri, A., (1999), "Does more intense competition lead to higher growth?", Discussion paper No. 2249, Centre for Economic Policy Research. Nickell, S, (1996), "Competition and Corporate Performance", Journal of Political Economy, No. 4.
(4) Preliminary statistics from Eurostat for the year 2000.
(5)The Swedish Competition Authority, (2000), "Why are Swedish prices so high?", The Swedish Competition Authority's report series no. 2.
(6) SNS Economic council (2002), Unlimited competition, SNS-förlag. Exposure to competition is here defined on the basis of concentration and other barriers to competition such as companies' actions, regulations limiting competition, entry and exit barriers.
(7) Price comparisons for housing are difficult. Differences in quality and structure between the Swedish and European housing markets makes accurate price comparisons difficult. For instance, there is a large supply of rental apartments in Sweden, while in other countries there are few of these and they are often greatly subsidised. Statistics Sweden estimates that the relative price for housing in Sweden is probably rather exaggerated.
(8) The Swedish Competition Authority, (2001), "Why are construction goods more expensive in Skåne and food cheaper in Western Sweden?", Swedish Competition Authority's report series, no. 1.
(9) The Swedish Competition Authority, (2001), "Can local authorities push down food prices?", Swedish Competition Authority's report series no. 4.
(10) Government Bill 01/02:167, "Changes in the Competition Act for more efficient combating of cartels, etc.".
(11) This applies if one compares with other profit-making banks. The non profit-making banks, for instance, co-operative banks, which are common in Germany and France, have even lower profit margins. The margins are calculated as the difference between the interest rate offered to consumers and the interest the bank receives if they invest the money without risk, e.g. in treasury bonds.