Once again, the prospects of lower growth and weak recovery are making headlines after the IMF and World Bank meetings in Washington. This time it’s the IMF that’s revising long-term growth prospects. Weak productivity and an ageing population mean that we might see low growth for an extended period of time.

Having spent three days in New York and five days in Washington talking to ministers of finance, central bank governors, IMF staff, macroeconomists from investment banks and hedge fund traders, the message is clear: structural reforms are necessary in all major economies. We urgently need to increase labour force participation and boost productivity.

But I’m struck by how difficult these reforms are. Most people agree that the US needs to invest more in education. Publicly financed pre-schooling, for example, could improve human capital and limit income inequality in the long run, when productivity differences in the labour force are compressed. Childcare reforms would increase female labour force participation. Over in Japan, there is an urgent need to open up the economy to internal and external competition. Labour market reforms in Europe could lower unemployment and increase growth. All of this is common knowledge among economists, but still very little progress is made. The political economy of structural reforms is difficult, very difficult indeed.

I’ve spent a lot of time thinking about structural reforms and also implemented a fair number of important policy changes during my eight years as Sweden’s minister of finance. Between 2006 and 2014, the government cut taxes for low- and medium-income earners to below the average for OECD countries. It also reduced the overall tax burden on labour from 48% to 43% of GDP, reduced the corporate tax rate from 28% to 22% and abolished the wealth tax. Unemployment benefits were reformed and requirements for early retirement and sickness benefits were tightened. The government also made it easier for those over 65 to postpone retirement and continue working, reduced barriers to labour migration, introduced more competition in healthcare and elderly care, deregulated closed sectors such as pharmacies, and privatized important state-owned companies. In 2010, at the height of the global crisis, our government was re-elected with increased support, but in 2014 we lost the election.

Based on this experience, here are five concrete lessons on how structural reforms can be implemented and pave the way for future electoral victories. I will also try to explain why we lost the election in 2014 (here’s a clue: we didn’t follow the reform strategy that I will outline).

  1. Front loading

The most important lesson when it comes to putting in place structural reforms is that they should be front loaded. Most reforms that increase labour supply, reduce threshold to the labour market, reduce regulatory barriers, or open sectors to more international competition have different long- and short-term effects. In the short run, there will be losers. Those who have worked in protected industries will see lower wage premiums. More labour market flexibility will reduce job security and increase turnover. Those who feel these negative effects are likely to voice their discontent. But in the long run, higher productivity means better real wages and increased labour supply, which will eventually increase the employment ratio and reduce social exclusion. The medium-term gains for society as a whole might offset the short-terms costs for certain groups.

The political implications are clear. When structural reforms are implemented close to an election, the short-term impact will dominate the debate, and the more nebulous long-term gains will be written off as uncertain forecasts. Gerhard Schröder is maybe the clearest example of this. Between 2003 and 2005, the SPD government cut taxes, tightened unemployment benefits, introduced local exceptions to the wage bargaining system to prevent job losses, increased pressure to participate in active measures, and created “mini jobs” to increase employment opportunities. Unemployment in Germany fell from 12% in 2005 to 5% in 2014, probably one of the most successful turnarounds in Europe. Schröder lost the election in autumn 2005. Had he initiated the process directly after the 2002 election, might he have been re-elected?

Front loading is also important because of how structural reforms impact public finances. If you cut taxes for low-income earners and finance it through tighter welfare benefits, increased labour force participation will reduce expenditure. That’s because fewer people will enter the welfare system and tax revenues will increase as those remaining in the labour market continue to pay taxes. Given that standard models for forecasting public finances are static in the sense that behavioural changes are not fully taken into account, actual outcomes are likely to surpass forecasts. This means that a front-loaded reform programme creates unexpected fiscal space. If reforms are implemented directly after an election, there will be room to spend in the re-election campaign.

  1. Keep it simple!

The second lesson for structural reforms is the importance of robust and simple implementation. Political systems are complex. Implementation is decisive for the outcome. Any given economic problem can be addressed with a number of different policy proposals. Some are very simple to implement. Cutting unemployment benefits from 80% to 75% of a claimant’s previous income can be implemented in the magic first 100 days in office. Implementing stricter work search requirements for the unemployed would require the buy-in of bureaucrats on the ground. Any reforms that require extensive and complex legislative work are more likely to get stuck at committee phase. Any measures that rely on a cultural change among staff are more likely to fail. Any reforms that rely on local government for implementation risk getting bogged down in local politics or simply failing. Any strategies based on the idea that unions and other interest group will be constructive are more likely to fail.

So keep it simple! Change parameters that are under central government’s direct control, that don’t depend on local authorities and bureaucrats.

This means that those reforms that seem key – improving the quality of education by boosting the stature of teachers, increasing professionalism in healthcare provision, streamlining building regulations or urban planning processes – are also those that are most likely to fail.

  1. Remember Keynes

The third lesson is that for a programme to be successful, it has to have a strong Keynesian component. The short-run economic dynamics have to be taken seriously.

If we look at past successful structural reform programmes, or if we take the example of countries that have had huge turnarounds, there has almost always been some form of demand-side stimulus. Given that reforms tend to come when economic problems are too obvious to be denied, fiscal space is often lacking. It’s therefore always helpful, if not indispensable, to increase demand for reform through monetary policy and the exchange rate. It’s sometimes argued that a weakening of the exchange rate or more expansionary monetary policies reduce pressure to reform. I beg to differ. Most structural reforms are easier to implement and less costly in social terms if there is strong demand.

The opposite is also true. A major currency depreciation without structural reforms might bring temporary relief, but it will also increase import prices and inflationary pressures. When a major depreciation is followed by structural reforms and increased labour supply, a sustained period of strong growth is more likely to follow. The fact is most successful structural reforms coincide with currency depreciations of around 20% to 30%. This is an important message to inflation hawks in the Eurozone: there can be no successful programme of restructuring unless monetary policy and exchange rates are supportive. The economic policies I implemented in Sweden were no doubt well balanced, but a hefty depreciation of the krona between 2008 and 2010 didn’t hurt.

  1. The camel in the room

It is easier for a camel to pass through the eye of a needle than for a rich man to enter the Kingdom of God. This could be translated into a key lesson for reform strategies. It is absolutely necessary to take equality into account. It is no coincidence that structural reforms are more popular in political parties whose supporters are better educated, live in urban areas, work in the private sector, own their own homes along with other assets, and don’t need to rely on benefits. It also shouldn’t come as a shock that political parties whose supporters are less educated, work in protected segments of the economy and rely on benefits will resist reform. This is not primarily about class or ideology: it is logical to resist policy proposals that will cause large immediate losses for uncertain gains in a distant future.

Structural reforms that boost productivity will improve average real wages, at least in a well-functioning European labour market with strong social partners. Shanta Devarajan, a brilliant economist at the World Bank, has often spoken about pro-poor reforms. We need to be careful when choosing what type of reforms we implement. Some reforms are more likely to have a positive impact on income distribution. If you get an equal GDP effect from either cutting the highest marginal tax or introducing an earned income tax credit for single mothers, choose the latter.

If voters are going to accept structural reforms, it’s also important to be upfront when dealing with income distribution. According to an almost proverbial saying from an editorial in The Economist: elections are decided when either responsible broadly working class groups can’t stand voting for irresponsible left-wing policies, or the compassionate middle class cannot stand further increases in income inequality.

If you want to implement structural reforms, the Gini coefficient should be a benchmark to keep in mind. If front-loaded reforms create fiscal space, use it for redistribution. If you need to cut taxes for the rich, as we did with the wealth tax in 2007, make sure there’s no election coming up, and try combining it with other measures that redistribute wealth.

  1. May the force be with you

Entrepreneurship is the strongest transformational force in society, and it will be even stronger in the future. In an ever-changing economy, every society needs to mobilize an army of entrepreneurs. Successful structural policies should release this force and direct it towards the good of society.

But that’s easier said than done. For many reasons, it’s very difficult to encourage entrepreneurship through policy changes. There is always a risk that policy proposals, particularly if influenced by special interest groups, will only benefit existing and sometimes stagnating companies, rather than driving the creation of new and dynamic ones. Some of the most important instruments, such as lower marginal taxes, reduced taxation on stock options or wealth tax cuts, are likely to be bad for the Gini coefficient. It is therefore important to embed measures that achieve cohesion into political proposals for increasing entrepreneurialism. Taxes and expenditures can be used to spread the gains of increased productivity to many, if not all, groups.

This is my take on structural reforms. Prepare to front load, keep it simple, remember Keynes, deal with the camel and release the force of entrepreneurship.

But this raises an obvious question: if this recipe works, why am I not minister of finance anymore?

The 2014 election campaign focused on high youth unemployment and a lack of progress integrating refugees into Swedish society. Although overall employment figures were good, youth unemployment was high: while unemployment levels for those aged between 25 and 65 were on par with Germany, our youth unemployment figures were closer to those of France. This problem has existed for the past decade so it is not transitional.

One alternative would have been to launch a new round of reforms after the victory in 2010. With the benefit of hindsight, this seems the obvious conclusion. Increasing labour market flexibility to reduce thresholds for youth and immigrants would have helped.

So why didn’t we do this? It was largely down to the very strong growth in 2010 and 2011. GDP grew by 6% in 2010 and 4% in 2011 (later revised down to 3%). It was natural to believe that growth would remain strong for years to come. The complacency was reinforced when opinion polls in 2011 and 2012 showed strong support for the government.

Growth in jobs creation and a rapid increase of labour force participation also suggested that the supply-level policies would eventually pay off for both immigrants and young people. This turned out to be a mistake.

Then the situation in Greece kicked off. Europe entered into another round of crises. This time the Riksbank, Sweden’s Central Bank, was constrained by household indebtedness and severely underestimated the impact on labour market reforms and increased labour supply. The European Central Bank did not provide support in dampening the crises (quantitative easing clearly should have been put in place in 2011) and there was no European fiscal expansion. Sweden was alone in a very weak environment.

Tight monetary policy also made the krona too strong, which dampened job growth in the export sector. As there was no fiscal policy response in either 2011 or 2012, growth in 2012 and 2013 was on the low side, and the output gap remained large and negative. When the Greek crisis returned, our initial fiscal policy response was too weak. We recognized this, and in the budget proposals for 2013 and 2014, we included substantial stimulus measures. There was a new high-speed train, along with other infrastructure investments, further reductions in corporate taxes, more education spending, increased active labour market measures, reforms to streamline urban planning and building regulation, and tax cuts for low- and medium-income earners and pensioners. It was essentially a good policy mix to increase long-term growth and boost short-term demand.

It is, all in all, clear that we committed a couple of serious mistakes. First of all we didn’t front load. Instead, we actually back loaded the reforms, so implementation was back-to-back with the election.

Furthermore, we forgot about the importance of Keynes. Demand-side stimulus was too weak; we should have accepted larger deficits in 2011 and 2012, given all the supply-side reforms and the ongoing crises in the Eurozone.

We also made a third mistake: we didn’t keep it simple. To deal with youth unemployment and the problem of integrating immigrants, the government entered long negotiations with partners on the labour market. After some 250 meetings and no results, the conclusion are clear. The unions will not voluntarily create a youth apprentice system because that would imply lowering entry wages for young people. There is no willingness among union leaders to accept more flexible hiring rules, which is essential if we are to improve resources to re-educate people and ensure they have the skills needed for tomorrow’s job market.

It was clearly a fatal mistake to not move forward with reforms to employment protection legislation and other measures to increase labour market flexibility. This is the most obvious reason the government lost the election in 2014.

If we as a government had implemented a comprehensive programme of reforms as soon as we were re-elected in 2010, we might have had a better chance in 2014, particularly if we had also used both fiscal policy and a growth-boosting programme in 2012 and 2013.

The lessons from our mistakes are clear: front load, keep it simple, and do not forget Keynes. Reform policy is tough. And it’s easy to make mistakes, such as getting complacent on the back of positive opinion polls and an exaggerated belief in your own policies.

But our achievements – GDP growth higher than the US, the UK, Japan and the Eurozone, higher investment levels, jobs creation, healthy public finances, growth in real wages and an election victory in 2010 – show that these five lessons are worth remembering. A lesson, sadly enough, also reinforced by our mistakes.

Author: Anders Borg, former Swedish finance minister, Chair of the World Economic Forum’s Global Financial System Initiative

Image: A trader looks up at a chart on his computer screen while working on the floor of the New York Stock Exchange shortly after the market opening in New York July 11, 2013. REUTERS/Lucas Jackson