McKinsey Quarterly

Reforming the public sector in a crisis: An interview with Sweden’s former prime minister

| Interview

Government leaders around the world face a daunting dual challenge: they must control and, in the long term, slash major budget deficits fueled by the economic crisis while at the same time improving the performance of the public sector so that it can meet its complex and ever-rising obligations.

Former Swedish prime minister Göran Persson is no stranger to that challenge. Even his political foes recognize his achievement.

In the early 1990s, Sweden suffered its deepest recession since the Great Depression. Although the Swedish crisis was homegrown, its causes and effects resemble the events unfolding in the world today. After years of strong domestic growth driven by easy credit and high leverage, a real-estate bubble burst, leading to the collapse and partial nationalization of the banking sector. Domestic demand plunged as the household savings ratio soared by 13 percentage points. In three years, public debt doubled, unemployment tripled, and the government budget deficit increased tenfold, to more than 10 percent of GDP, the largest in any OECD1 country at the time.

Persson was appointed finance minister when the Social Democrats returned to power, after the 1994 elections, and became prime minister two years later. In order to regain the confidence of international lenders—and so pave the way for stability and sustainable growth—he knew that Sweden had to reduce its budget deficit dramatically. It took four years for the Swedish government to balance its budget. By 2006, when Persson and his party lost power in the general elections, the country had almost halved its public debt, to just above 40 percent of GDP.

Göran Persson recently spoke with McKinsey’s Alastair Levy and Nick Lovegrove about what it takes to put troubled state finances in order and, at the same time, to improve the way the public sector works.

The Quarterly: What is the prerequisite for implementing a successful crisis program?

Göran Persson: The electorate must understand that drastic measures are required. A crisis program will hurt, and you will need a mandate from the voters if you are to succeed. This makes it difficult for an administration that is in power without such a mandate to take the lead. But it is a fantastic chance for the opposition, provided that there is broad awareness of the gravity of the situation. My party was elected in 1994 because we promised to carry out the harshest program with the deepest budget cuts and the sharpest tax increases.

The Quarterly: What advice would you give incumbent leaders who don’t have a mandate from the voters for instituting radical reform?

Göran Persson: You have to make it absolutely clear that you are putting your office at stake; that you are prepared to call new elections or, if your parliamentary group is not behind you, to resign. The forces working against a harsh crisis program are very strong—almost every area of the public sector has its own vested interests—so any sign that you might waver in your commitment will doom the program to fail.

The Quarterly: Please summarize the lessons you have learned about leading, designing, and implementing the process for putting state finances in order.

Sweden’s budget and reforms under Persson

Göran Persson: First, it is extremely important to be in the driver’s seat. You must make it clear that you are responsible for the process and that you are prepared to put your position at stake. Second, the consolidation program must be designed so that the burdens are shared fairly. Public-sector cuts will hurt the most vulnerable people in society, so those who are better off need to contribute—for example, by paying higher taxes. Public support for tough policies would quickly deteriorate if they were not perceived as fair, and parliament would lose the political will to make hard decisions. Third, the consolidation program has to be designed as a comprehensive package; if you are in as deep trouble as we were, an ad-hoc hodgepodge of measures will only have a limited chance of success. Moreover, by presenting the measures together, it becomes clear to all interest groups that they are not the only ones being asked to make sacrifices. It also has to be a front-loaded program. By starting with the most difficult measures, you demonstrate your resolve and increase the chances of achieving the early results, which will be important for getting the continued support that is critical for sustaining the effort.

Transparency is the fourth lesson. You must never play down the effects of the program’s measures. On the contrary, remind the public again and again that this will hurt. It is one thing to get support in parliament for the program; it’s another to stay in control during the implementation phase, when the measures become real for ordinary people in their daily lives. You must also be completely honest when you communicate with financial markets. Clarify assumptions and calculations. Don’t use any bookkeeping tricks. Only then can you recover credibility; only then can the program earn legitimacy. Indeed, you should always go for conservative estimates. If, for instance, you estimate that economic growth will be 1.5 percent and you end up with 2.5 percent, you will have solved much of the credibility problem.

The Quarterly: The electorate’s patience is never endless. How much time do you have until it runs out?

Göran Persson: You have two years. If you are not in command of the process by then, you will lose momentum and soon face the next election—where you will be replaced. We survived the 1998 election and were rewarded politically for what we had done by being reelected once more in 2002, when the good times returned and we were in firm control of the public finances.

The Quarterly: Cutting the state budget during a crisis puts pressure on the public sector at a time when its services are perhaps more important than ever. How did you handle this problem?

Göran Persson: Restoring the health of our public finances was the prerequisite for preserving the Swedish public sector in the long term, and this would not have been possible without sacrifices. One-third of our program consisted of tax increases, and two-thirds of spending cuts, both in the operational budgets of the central and local authorities and in the legislated levels of welfare transfers. We cut pensions, sick-leave compensation, and unemployment benefits, which hurt people who already had only small margins in their household finances. That shouldn’t have been necessary in an ideal world, because lower welfare transfers reduced domestic demand and tax revenues and thus had a negative impact on growth and employment and a small net effect on the budget. But we had no choice. High interest rates made it necessary to regain the confidence of investors all over the world whose perception was that Sweden’s generous welfare model was to blame for the crisis. In fact, it wasn’t until we cut unemployment benefits and got into open conflict with the trade unions that market interest rates started coming down.

The Quarterly: It’s often said that with a crisis comes an opportunity for reform. Did you use this opportunity to improve the long-term performance of the public sector?

Göran Persson: Yes, the cuts in government consumption became a driver of improved efficiency, since public authorities were forced to do the same job on unchanged or reduced budgets.

In addition, we pursued targeted policies with various objectives. One strategy—aiming to improve productivity, service quality, and freedom of choice—involved the liberalization of telecommunications, mail, railways, and other infrastructure industries. It also involved allowing privately run providers to compete with public ones in providing tax-financed services for the school system, health care, child care, and care for the elderly.

Another measure was to introduce information technology to broad layers of the population through a tax-deduction scheme that allowed workers to obtain a home computer under a favorable leasing agreement with their employers. The penetration of IT in Sweden during these years outpaced every other country in the world, which made it possible for authorities like the Tax Agency to go online at an early stage. Indeed, I’m quite confident today that information technology improves government productivity as well as the delivery of its services. More and more of the communication between Swedish public agencies and citizens now takes place on the Web, and many Swedes do their annual tax submissions over the Internet, allowing for a very efficient processing of taxes. I think our tax agency is one of the most efficient in the world and very much so because we are using modern technologies. We have one of the world’s largest public sectors and, along with the Danes, the world’s highest taxes, claiming almost 50 percent of GDP. We are also very good at collecting these taxes.

A third strategy was to give people with basic schooling the chance to complete a secondary education that would qualify them for university studies.2 It was a straightforward system: an employed worker would get the equivalent of the unemployment benefit if he or she entered an adult-education program and if the employer agreed to replace him or her with an unemployed person. The employer’s cost was unchanged, and the state’s cost was limited to the education itself. Believe it or not, more than 10 percent of the workforce seized this opportunity between 1997 and 2002. It was mainly women who did so, and many went on to study at a university. When the business cycle turned up again, they became a very good resource on the labor market, not least in the public sector. This education scheme served a dual purpose: it eased the pain of unemployment and increased Sweden’s long-term competitiveness by lifting the average competence level of the workforce.

The Quarterly: What approach did you take to set efficiency targets and drive savings across the government?

Göran Persson: We introduced three-year ceilings on public expenditure for each ministry. Within this ceiling, we gave the ministries and public agencies some flexibility to distribute their expenditure levels between the years in each three-year frame as long as they reached their final target. These caps on expenditure were the main driving force. Sweden has a decentralized system of government, so even though we set the guidelines it was up to each authority to figure out how to fulfill its service obligations while still achieving the required spending cuts. The budget cuts for the authorities and agencies amounted to a grand total of 11 percent from 1995 to 1998. After that, we built in an efficiency factor based on productivity in the private-service sector, which the public agencies had to match. By doing so, we continued to put pressure on them to improve their efficiency and produce more or the same for less. The result was that they started to examine expenditures that they had regarded as impossible to influence—for instance the location and rental cost of their offices—and they also became more careful about whom to employ and about developing the staff they already had.

The Quarterly: As the political leader, what was your experience with trying to get the civil servants on board and making them partners in the initiative?

Göran Persson: They had never experienced a crisis of this magnitude. Some reacted to it as a professional opportunity to perform a very significant task. Others felt betrayed by the cuts and that it was not their role to deal with productivity or efficiency issues. In the end, though, it was quite easy to get the civil servants on board because they were all conscious of the crisis and its dangers.

The Quarterly: Did you make many personnel changes, particularly in important positions?

Göran Persson: Only gradually and in a small way. It’s very easy to get rid of people, but it’s difficult to find new ones that you can be sure are better. So I find that it’s often wiser to stick with the staff you have. It is, after all, the politicians who are responsible for restoring order in the country’s finances, so it’s up to them to lead, support, educate, and stimulate those who carry it out. Sometimes you are successful in this regard; sometimes you fail.

The Quarterly: Did you set up some kind of machinery at the center of government to monitor departments and agencies and to intervene, when necessary, to move things along?

Göran Persson: No, we did not. We were in such acute crisis that we had to move as quickly as we could, so we executed the program without reflecting in detail on its implementation. We monitored two indicators very closely; one was the bottom line of the state finances and the other Sweden’s interest rate levels, because financial markets reacted very quickly to the program and its progress. If I had to do it all again, I would perhaps set up some centralized unit just to monitor progress and to spread ideas and best practices.

The Quarterly: Did the process lead to significant changes in the way government worked and the way it developed and delivered its services?

Göran Persson: The efficiency targets had positive consequences for public services, at both the state and local levels. At the local level, the targets encouraged public agencies to collaborate, leading to better services for the people. Similarly, as a response to the remit we gave government agencies—such as the tax and social-security authorities—to improve their efficiency to private-sector levels, they started talking to each other and cooperating more closely than before. We didn’t plan these changes, but they were positive nevertheless.

The cabinet was another example of change. People tend to view it as a tight-knit team, but it is not. Ministers are constantly competing with each other for the available resources. This was not the case during the crisis. In fact, it was the only time in my 15 years as a cabinet member when I felt that I was leading a real team where everybody was prepared to contribute and to help each other. Why? Because we all understood that the budget deficit, if left unchecked, could destroy the public sector as we knew it. We also knew that beating the crisis required us to work as a team, because if just one minister leaks to the media that his or her area of responsibility is carrying an unfair share of the burden, the whole process will soon break down. You must realize that the cabinet is one thing; the parliament, however, is something else, and you can never take the support of your parliamentary group for granted. If there is the slightest dissension between your ministers, their support groups in parliament could block bills that you are bringing to the assembly. This would be very serious. A budget-consolidation process of this kind requires not only a state budget: the budget needs to be followed by perhaps 50 or 100 different initiatives that all have to pass through parliament. So if you cannot keep your team together, you will find yourself on a very slippery slope.

The Quarterly: What levers did you have at the center for influencing change at ministries that were not making good on their efficiency targets?

Göran Persson: Each ministry had its own bottom-line target, and if it didn’t make good on that target there would be a discussion with the ministries’ top managers. Where needed, I or my finance minister became directly involved in discussions with departmental ministers. In doing so, we suggested ways to move forward, but we would never tell them what to do. Giving direct and detailed orders would have broken the internal ethics of the budget-consolidation process—which we had agreed to achieve as a team. It would also have given the finance minister or prime minister ownership of somebody else’s task.

In fact, what is taught in the private sector about the importance of building well-functioning top teams applies to government as well, except that it’s harder in government. Much more transparency is required, and every little detail can become public knowledge. Moreover, your ownership of the process is under constant threat from the opposition and, perhaps, your own parliamentary group. This makes it essential to build loyalty and solidarity within your team of ministers or else you won’t achieve anything. Political leadership is often said to be about visions and ideas. But it is also about ensuring that a transparent public organization can achieve productive results in its daily work not only once but again and again every year, and under constant external pressure.

Explore a career with us