The UK and the single market By John Springford
This article was submitted as evidence by the Centre for European Reform for the review of the balance of competences between the United Kingdom and the European Union (Internal market)
What are the essential elements of an internal market and against what criteria should we judge its economic benefits? How deep does it need to be to be effective? There is no threshold beyond which the removal of trade barriers becomes ineffective, at least in economic terms. Barriers to trade are numerous, and eliminating them is a potentially limitless process. Thus, a truly internal market may only arise in a highly centralised nation-state. Once countries have cut tariffs to zero, import quotas can go, then goods safety standards can be harmonised. Free labour and capital movement allow the two main elements of the production process to move to the places where they may be most productively deployed. A completed single labour market would have common qualifications, pensions, and unemployment insurance. Tax differences distort trade – especially consumption and corporate taxes – and so a ‘completed’ internal market would harmonise tax rates. It is more useful to think about the European single market as a continuous bargaining process between member-states, who want both the growth in trade that arises from integration and also regulatory sovereignty – but must choose. The degree of integration reflects how far nation-states are willing to go. Negotiations between nation-states will not arrive at a magic formula that perfectly balances national regulators’ knowledge of local markets and firms, democratic accountability, and trade opening. Trade-offs and deals, based upon memberstates’ perceptions of their interests, predominate. There is, no doubt, poorly drafted and economically costly regulation that emanates from Brussels, and some of it is more illiberal than Britain would choose. Much employment legislation probably imposes more costs than it confers benefits. But the single market’s bargaining process means that losses must be set against gains. No ‘Goldilocks’ formula for a perfectly functioning market is available.
Has the bargaining process delivered much trade? And at what regulatory cost? Unfortunately, an accurate cost-benefit analysis of single market legislation is impossible. There are over 3,000 single market directives and regulations in force, each of which gives benefits to producers, workers, or consumers, as well as imposing costs on them. European Court of Justice rulings add to the body of single market law. And its costs and benefits vary by member-state, because national governments are free in many cases to strengthen EU regulations, by adding rules as they are written into national statute. Some have attempted to quantify the costs incurred by the British economy by EU regulation. We should be very wary of these estimates, for two reasons. First, they are based on the UK’s impact assessments, which mostly do not put a number on the benefits. Second, they do not evaluate the counterfactual: would regulation imposed by British authorities be less costly? However, various empirical analyses have been conducted to examine how much extra trade the single market elicits. One is the ‘gravity model’, which establishes how much trade one might expect between countries, given the size of their economies, their distance from one another, and other factors like a common language. If EU countries trade with each other more than the expected amount, this means the EU’s single market and, to a lesser extent, the euro are responsible. The UK Treasury estimates that the EU’s internal market as a whole – customs union, four freedoms of movement, and removal of non-tariff barriers – has boosted trade between EU members by 38 per cent of GDP. The Treasury researchers found a much smaller impact on Britain than for the EU as a whole – it led to an increase in trade of around 7 per cent of GDP.1 The European Commission estimates that the single market programme from 1992 produced
1: UK Treasury, ‘EU membership and trade’, 2005. .
The UK and the EU single market March 2013
info@cer.org.uk | WWW.CER.ORG.UK
1
around 2 per cent growth in EU output. Facing greater competition, companies cut margins by around 1 per cent. Productivity in labour, capital and land use increased by half a percentage point.2 Other estimates offer similar results. The economist Carsten Fink found that services trade was one-third higher in the EU than elsewhere.3 Yet trade is not the only measure of integration. Trade is good in itself, as it gives buyers more choice of cheaper or different goods. But trade also puts competitive pressure on indigenous firms, and gnaws away at profits, putting pressure on managers to use capital and labour more productively. One way to measure this process is through price convergence: weak competition leads to
divergences in prices across the EU, as incumbent firms can sell at high prices without fear of losing market share to foreign firms. Price dispersion has been falling quickly since 2004 in the accession member-states in Central and Eastern Europe, as they have integrated with the west. But prices have been converging much more slowly in the 15 western and Nordic member-states, falling from an average variation of 14 per cent in 2001 to 13 per cent in 2010.4 A related indicator, the ‘markup’ difference between firms’ costs and the prices they charge, shows that the single market in goods is far more integrated than that of services. Markups in manufacturing have fallen in the last two decades, while those in services have grown.5
To what extent is EU action in other areas – for example, environment, social, employment – necessary for the operation of the Internal Market, as opposed to desirable in its own right? If one sees the single market as a political bargain, as opposed to an economic policy designed by technocrats, then these actions are necessary to settle a deal. Some EU environment, social and employment policies may not be economically necessary, but act as a political quid pro quo for market opening. These policies arise from some member-states’ fears of social and environmental ‘dumping’. They argue that firms in less-regulated member-states may receive a competitive advantage with more open trade, because the firms’ costs are lower if they pollute the environment or fail to protect their workers. The single market operates by creating trust between national governments, businesses, workers and regulators. French trade unions, for example, fear that producers based in Central and Eastern Europe will enter French markets, taking French jobs away, because labour costs are lower. So, they demand a social and employment ‘floor’ guaranteeing a minimum level of labour rights. Similarly, some member-states fear that without strong anti-pollution rules, those countries with less onerous protections would have an advantage – they could use cheaper, but more polluting energy sources, for example. These minimum standards allow countries to accept the free flow of goods and services, people and capital, with less fear that liberalisation will lead to lost jobs and falling wages for some, as
production shifts to countries with weak labour and environmental protections. From a purely economic perspective, most economists have found that global growth in trade has had a small but detrimental impact on wage inequality, which has been growing throughout the developed world since the 1980s. But it is unlikely that the single market had much to do with this. Trade with countries with cheaper labour or more lax regulations has an impact on lessskilled workers at home. But the majority of trade in the EU is conducted by the rich countries in the west of the continent, which have similar levels of development, and labour and environmental protection – compared to the difference between Britain and China, for instance. Moreover, the impact of trade on inequality is small, compared to technological change: workers’ skills have been replaced by machines and software, putting far greater downward pressure on their wages than lowcost production in other countries. This suggests that environmental, social and employment protections at the level of the 27 are not really necessary for the single market to improve most people’s lot. However, the rules are not especially onerous. The EU demands no minimum wage. And it is likely that many of the regulations protecting air, land and water would be replicated by the UK if it left the EU.
What is the right balance between harmonisation and mutual recognition? What evidence is there that harmonisation has worked well or badly? Do EU regulations encourage trade between its members, while tying up Britain’s domestic economy
in red tape? This argument would be persuasive if two conditions are met. First, it would hold if EU institutions
2: European Commission, European Economy Economic Papers, No 271, January 2007. 3: Carsten Fink, ‘Has the EU’s single market program led to deeper integration of EU services markets?’, Sciences-Po, 2009.
4: John Springford, ‘How to build European services markets’, CER policy brief, September 2012. 5: Harald Badinger, ‘Has the EU’s single market programme fostered competition? Testing for a decrease in markup ratios in EU industries’, Austrian Central Bank, 2007..
The UK and the EU single market March 2013
info@cer.org.uk | WWW.CER.ORG.UK
2
aligned member-states’ regulations by making them identical – so called harmonisation, rather than allowing states to recognise each others’ regulations – or ‘mutual recognition’. Second, if, in the process of aligning national regulations, EU institutions made rules stronger, leaving Britain’s economy more heavily regulated. Is there too much harmonisation? Some single market legislation takes the form of regulations, which have direct effect, and replace all pre-existing rules on member-states’ books. But most EU trade is conducted under the aegis of mutual recognition or directives (which member-states must implement, giving them some leeway, and which leaves them to enforce the rules). While this means that the majority of new regulation probably comes directly from EU institutions, or is enacted by member-states in response, it also leaves a fair degree of discretion to member-states. One-fifth of goods are traded under the mutual recognition principle, without the need for a directive.6 The majority of the other goods markets are governed by so-called common objectives – whereby the EU sets common principles and minimum standards, and member-states are free to vary rules within bounds, while allowing goods produced abroad to enter their markets without restriction. Services, which make up 70 per cent of economic activity, are still largely regulated at the national level (although the 2006 services directive made memberstates get rid of rules that unfairly hampered foreign businesses from entering their markets). Member-states still have broadly discretionary regulatory powers in public services, health and safety, environmental regulation, labour markets and professional qualifications. Moreover, it should not be taken as axiomatic that harmonisation is a bad thing – or just a way to create a minimum floor of regulation so that mutual recognition can operate. Where consumers need a product to be safe and to do what its manufacturers say it will, it makes sense for the EU to issue harmonised rules so that companies across the continent compete on price and quality – and the consumer benefits. Various studies have shown that harmonised standards raise trade in manufactures.7 This may be because they become more trustworthy. Standards can also reduce costs for producers. While they may have to change the product to meet the standard, they save on consumer research
and marketing, to work out what level of quality foreign consumers are used to. In services markets, the mutual recognition principle could be deployed to a greater extent. The services directive removed many national regulations that stopped foreign companies from entering. But many remain: the owners of German accountancy firms, for example, must be the accountants themselves, which makes it difficult for foreign companies with different ownership structures to enter German markets. And many member-states have long lists of regulated professions, and do not recognise other nationals’ qualifications or, if none are required in other countries, their experience. However, opening EU services markets will also need more harmonisation in some sectors, such as retail financial services. Consumers cannot appraise the quality of the service at the point of signing up to a financial product. Harmonised rules about consumer information, what products can and cannot be sold, and how consumers are paid back if the service is poor will be necessary for deeper integration to be achieved. Does the EU impose more burdensome regulation than the UK would choose? Many rules do get pushed through the Council and Parliament with the single market label, simply because single market decisions are taken by qualified majority voting, and do not require unanimity. Social and employment law, such as the working time directive, the part-time workers directive and the temporary workers directive, do increase the regulatory burden in the UK. Britain would probably not give workers the right to demand restricted hours, because they could simply refuse to take the job in the first place. (But they are currently able to opt to work for more hours if they would like.) Likewise, it would probably not offer such strong employment rights for part-time and temporary workers. However, the number of part-time and temporary workers is growing quickly, as economic stagnation continues – which suggests that the cost of the directives is not insurmountable for employers. More broadly, if EU regulation were onerous, the OECD would not rank the UK as one of the least regulated developed economies. In product markets, it is the least regulated economy in the OECD. Its labour markets are third-least regulated.
Why is the internal market so much deeper in some areas than others? How effective has implementation of the internal market been? The EU has integrated faster than the rest of the world since the single market programme began in 1992. However, it still has a long way to go to match the degree of integration achieved by United States – the continental market par excellence. Trade between the
American states, as a percentage of their GDP, is 70 per cent higher than the EU-15. In 2009, the EU-15 traded 55 per cent of their output, while the American states traded 93 per cent.
6: Jacques Pelkmans and Anabela Correia de Britto, ‘Enforcement in the EU single market’, CEPS, 2012.
7: Marc Vancauteren and Daniel Weiserbs, ‘Intra-European trade of manufacturing goods : An extension of the gravity model’, 2005..
The UK and the EU single market March 2013
info@cer.org.uk | WWW.CER.ORG.UK
3
That difference is down to the US’s much more integrated services market. The large majority – three-quarters – of the growth in trade within the EU has come from rising trade in goods. There is still significant ‘home bias’ in services: citizens of the EU-15 buy 94 per cent of their services from firms based at home. Services make up 70 per cent of the EU’s output and employment. And they are becoming increasingly tradable: the internet allows retail, for example, to take place without buyer and seller coming together. The same is true of design and architecture, engineering services, advertising, accounting and consultancy, among other sectors. The 2006 services directive sought to remove national restrictions on services trade and foreign direct investment. But it did not go far enough. Across the EU, the average number of restrictions on establishment or cross-border provision fell by a third – the majority of barriers still remain.8 The directive left national regulators with too much discretion in defining services trade barriers. This has meant that the total reduction in barriers to entry has not been as large as the authors of the directive hoped. In network services such as energy, broadband internet and transport, integration has been slow.9 These sectors need cross-border physical infrastructure for a single market to operate. For example, a European energy market requires major investment in crossborder interconnections, as well as backup storage for renewable power: if electricity companies in Germany generate surplus power, they can only sell it to Poland if enough electricity lines cross the border. A costeffective renewables system would need solar power in the Mediterannean region to be transported north, and wind and wave power transported from Northern Europe to the south, as both energy sources are intermittent.10 In order to create a European electricity grid, the Commission estimates €1 trillion of investment is needed. But the EU’s capital commitment is unlikely to be enough. The ‘Connecting Europe’ investments were cut from the proposed figure – €50 billion – to €29.2 billion in February 2013’s EU budget deal.11 The EIB’s project bonds for infrastructure investment will bring another €4 billion.12 The EU hopes that this public money will bring in private capital, but member-states have different ownership structures, systems of subsidy, and rules about infrastructure coverage, which act as a deterrent to investment in cross-border infrastructure. Alongside the single market’s partial construction, some problems with the enforcement of existing legislation remain. Much single market legislation takes the form of directives, and requires member-states to 8: John Springford, ‘How to build European services markets’, CER policy brief, September 2012). 9: Philip Whyte and Simon Tilford, ‘The Lisbon Scorecard X: the road to 2020’, CER report, March 2010. 10: Stephen Tindale, ‘Connecting Europe’s energy systems’, CER policy brief, October 2012.
implement them. Member-states often drag their feet on the implementation of directives, or ‘gold-plate’ them to make them stronger – and may in so doing set up barriers to trade. According to the Commission’s annual checkup of the single market, Sweden, Malta, and the Netherlands take longest to transpose new rules. Belgium, Poland and Italy have the largest backlogs of directives yet to make it into national law. Italy, Poland and France are the worst gold-platers. Member-states have also frustrated the Commission’s attempts to police national legislation to ensure it accords with EU law. The Commission has proposed that member-states hand over ‘correlation tables’ which show how much existing national rules diverge from EU directives, but has been rebuffed in some cases. National courts have defined various aspects of the e-commerce directive differently, which has forced companies seeking redress to take court action in several member-states at once. The ‘SOLVIT’ system gives regulators, businesses and people an informal resolution mechanism for difficulties when migrating and doing business in another country. SOLVIT centres in each country put businesses and people in touch with regulators, to get them to amend rules that constrain freedom of entry, or offer ways around them. But the numbers of businesses using the system has been disappointing. In 2011, less than 200 cases were resolved through SOLVIT.13 However, two reforms have improved implementation and enforcement, and could be built upon. First, directive 98/34 makes member-states submit any new regulation in goods markets to the Commission, which checks that it does not restrict imports from other member states. If it does, the Commission asks the member-state to amend the draft regulation. The Commission can also delay the national regulation for up to a year, which encourages national regulators to think about the impact on the single market beforehand, if they want speedy passage. The EU could consider extending this to services regulation, to prevent the gains that have been made from the services directive from being undone by new rules at the national level. Second, the European Parliament’s internal market committee has involved national regulators at an early stage in the legislative process, to tackle implementation problems before the directive comes into force. In the same spirit, the services directive was passed after national regulators evaluated each others’ rules: by comparing notes, and with a little peer pressure, regulators removed much that did not work or was unnecessary. Forming colleges of national regulators to help write major EU directives would be the obvious next 11: European Commission, ‘Conclusions on the Multiannual Financial Framework’, February 2013. 12: EIB, ‘EU-EIB Project Bond Initiative launched with start of pilot phase’, November 2012. 13: European Commission, ‘Making the single market deliver’, 2011.
The UK and the EU single market March 2013
info@cer.org.uk | WWW.CER.ORG.UK
4
step. It would help to head off implementation problems. And it would encourage national regulators to trust each
other, which would make it easier to pass legislation based upon the principle of mutual recognition.
To what extent has the UK kept requirements over and above the EU minimum, and what effect has that had on the UK’s place in the internal market? Have other member states done so, and if so with what consequences? The British government’s 2006 Davidson Review found no evidence that the UK tends to regulate more than the minimum. Previous studies that made this assertion counted the difference between the number of words in the original EU legislation, and the number of words that ended up on the member-state’s rule-book. But memberstates have different legal and regulatory systems, and so such differences are inevitable.14 As mentioned elsewhere, the OECD consistently finds the UK to be one of the most lightly regulated economies 14: HM Treasury, ‘Davidson review of implementation of EU legislation’, 2006.
in the developed world. The World Bank ranked Britain seventh in its most recent scorecard of the ease of doing business in different countries.15
John Springford
Research fellow, Centre for European Reform March 2013 15: World Bank, ‘Doing business 2013’.
To view all of our submissions to the review of the balance of competences, visit our website: www.cer.org.uk/publications/reviewcompetences
The UK and the EU single market March 2013
info@cer.org.uk | WWW.CER.ORG.UK
5