better public sector

Government by design: Four principles for a better public sector

By Diana Farrell and Andrew Goodman

Making progress on society’s biggest problems requires governments to make better use of data, involve citizens, invest in employees, and collaborate with other sectors.

Governments everywhere face a daunting paradox. On the one hand, they operate in an increasingly complex environment and must deliver on an expanded set of policy objectives. In a world characterized by macroeconomic uncertainty, rapid social change, and technological innovation, citizens’ expectations of what government ought to deliver are rising. On the other hand, governments are hampered by unsustainable debt burdens and shrinking budgets. The ratio of general government debt to gross domestic product for member states of the Organisation for Economic Co-operation and Development (OECD) now exceeds 100 percent.1 Meanwhile, public trust in government is eroding.2 Against this backdrop, not only must governments do more with less; they must do so in highly visible ways, if they are to regain the faith of their constituents.

The good news is that governments can deliver the performance their citizens need and expect—and, indeed, some have begun to do so. Based on the McKinsey Center for Government’s research into hundreds of cases of government innovation around the world, our on-the-ground experience working with governments, and numerous conversations with public-sector leaders and thinkers, we conclude that what works today is a more disciplined, systematic approach to solving public-sector management problems—in short, government by design.

Government by design calls on public-sector leaders to favor the rational and the analytical over the purely ideological, and to be willing to abandon tools and techniques that no longer work. Four principles are at its core: the use of better evidence for decision making, greater engagement and empowerment of citizens, thoughtful investments in expertise and skill building, and closer collaboration with the private and social sectors. Each of these principles is central to creating more effective yet affordable government.

The value at stake is staggering: prior McKinsey research suggests that improvements in government performance could amount to as much as $1 trillion in increased productivity and cost savings by 2016 in the G8 countries alone.3 Through government by design, public-sector leaders can move beyond partisan debates and politicized headlines, and make true progress on society’s most pressing problems.

Better evidence for decision making

Results-oriented governments are increasingly making use of hard data and statistical analysis to inform decisions. We have seen three forms of evidence-based decision making in government:

Collecting credible performance data

Governments must decide what to measure and how, always with an eye on the overall goal of the program or initiative. One of the goals of a government-wide transformation effort in France between 2009 and 2012 was to reduce the perceived complexity of dealing with the government. As part of this initiative, the government identified 50 life events—such as getting married or starting a business—during which citizens have to interact with public agencies. It then sought to simplify each of these interactions, all the while measuring citizen satisfaction to track whether the changes were actually working.4 Similarly, as part of a broader open government initiative, the city of Moscow is beginning to publish a dashboard of around 50 key performance indicators relating to the city’s health, education, safety, business conditions, and transportation outcomes. The dashboard acts as a scorecard for citizens, showing the city’s performance against these metrics.

Benchmarking consistently against peers

National and international benchmarks are powerful but underutilized as inputs into decision making, particularly in a world where governments everywhere face similar issues and no single government excels across the board. On almost any metric—from high-school graduation rates to unemployment to per capita health-care costs—there are wide variations in government performance across the OECD and even within large federal countries. An economist would say this inconsistency is to be expected since monopolies, absent any competitive pressure, tend to deliver suboptimal performance. But every government’s best practices can be useful to other governments and can motivate change. The introduction of the OECD’s Programme for International Student Assessment (PISA) study, for instance, caused a “PISA shock” that spurred reform in countries including Germany and the United States and subsequently highlighted a range of best practices in education (such as the building and nurturing of a high-quality teacher pipeline).5

Using data to design and improve interventions

Reliable, clean data can inform the design or refinement of government initiatives. The UK government’s Behavioural Insights Team was formed explicitly to use data about citizen behavior to improve the effectiveness of government interventions. The team sets up randomized control trials (long used in the medical field but only now gaining favor in the public sector) to test the impact of small changes, like adjustments in the language and tone of the letter that the tax department sends to delinquent taxpayers. In its first two years, the team paid for itself 22 times over in savings. It has identified interventions expected to save the UK government at least £300 million over the next five years.6 And the unit has started to advise other governments on how to use data and randomized control trials to improve government performance.

Evidence-based decision making creates real value, both financial and nonfinancial, for citizens. Done right, it allows government to assess policy and program effectiveness, measure progress, and engage in a more rational public debate on sensitive topics.

Greater engagement and empowerment of citizens

Innovative governments are making it easier for citizens to access public services. And the most forward-thinking governments are starting to master the shift from simply administering services to regularly engaging and empowering citizens, involving them in the design—and, in some cases, the delivery—of these services. This shift is not just about increasing choice and well-being; it’s also about boosting government productivity, with the help of technology and the use of open data.

Using innovative channels to make services more citizen-centric

The private sector’s responsiveness to customer demands has led to heightened public expectations of government. Because people can do their banking and shopping online, for example, they expect to be able to apply for driver’s licenses and submit tax returns online as well. Governments are investing to meet these expectations. The Estonian government’s e-services portal, visited by more than 10,000 users every day, allows residents to perform an ever-expanding array of tasks including applying for unemployment benefits, paying taxes, registering new companies, and even voting.7 But being citizen-centric isn’t just about the Internet: Australia, for instance, has pioneered mobile government offices—satellite-equipped trucks—that serve as a one-stop-shop for government services for people living in remote areas.

Soliciting citizen input to improve public services

Innovative governments are creating new ways for citizens to make their voices heard, giving them the ability to provide input into regulations, budgets, and the provision of services., one of the US government’s earliest e-government programs, allows citizens to search, view, and comment on federal regulations. Users post more than 27,000 comments on the site every month. Other governments are going even further to solicit citizen feedback: Iceland in 2010 chose 950 citizens at random to participate in the drafting of a new constitution, a significant example of “deliberative democracy” at work. And the city of Cologne, Germany, has used participatory budgeting: residents helped decide how to allocate a portion of the municipal budget.

Tapping citizens to help deliver better services at a lower cost

Citizens can play an important role not just in the design but also in the delivery of public services. New York City’s 311 system allows residents to report nonemergency complaints—about things like potholes or garbage collection—via a website, a mobile app, text messaging, Skype, or phone. Thanks to the Open311 platform, a free web-based application programming interface, the city now processes 60 percent of service requests online, lowering transaction and issue-resolution costs. Open311 platforms have been rolled out to other cities as well, including San Francisco and Chicago. These platforms, along with third-party apps such as SeeClickFix, empower citizens to do some of the work that has traditionally fallen to municipal employees: citizens in effect act as city inspectors. In a similar vein, the mayors of Boston and Philadelphia have each created an Office of New Urban Mechanics, which works with residents to fund and launch promising projects that address civic needs. Citizens—not government employees—come up with the ideas and do much of the work, but also reap the benefits.

The trend toward participatory government will only gain in strength. And by engaging and empowering citizens to codesign and codeliver public services, governments can not only better meet citizens’ needs; they can also shift some of the burden of accountability from the state to the people, allowing high-quality delivery of services in an environment of constrained resources.

Investments in expertise and skill building

Mission-driven employees are among the public sector’s most valuable assets. Unfortunately, many governments fail to get the most out of their people—they don’t invest enough in developing their employees’ skills and expertise. For instance, although government agencies have started to embrace “lean” principles such as value-stream mapping and Six Sigma process improvement, many are unable to sustain the impact from these initiatives because they haven’t been deliberate about building internal capabilities. In a 2012 survey of 974 public-sector leaders, we found that only about 39 percent of large-scale public-sector projects fully met their targets.8

Sometimes, the problem is that governments invest in the wrong kind of training. Research has demonstrated that adults learn six to seven times more through practice and feedback than through lectures, yet far too many public-sector training programs consist of classroom sessions or self-study modules.

Smart government institutions are ensuring that their employees develop and hone the skills that truly matter—whether those are core competencies, sector-specific capabilities, or broader expertise in strategy and risk management.

Using adult-learning practices to build core capabilities

Management skills are crucial to the success of any government program. A study by McKinsey and Oxford University revealed that more than two-thirds of budget overruns in large-scale IT projects are due to managerial—not technical—shortcomings.9 Best-practice government agencies are investing to make sure their managers are equipped with the requisite know-how. The US Department of Housing and Urban Development’s Office of Multifamily Housing Programs recently undertook a capability-building program that included a series of process improvements, the introduction of new managerial routines, and intensive coaching on problem-solving skills. The program yielded a reduction of more than 70 percent in the agency’s backlog of housing applications and a 35 percent productivity improvement. When Germany’s Federal Labor Agency undertook a similar program, the agency’s “customers” benefited: their average duration of unemployment fell from 164 days to 136 days. The Swedish Migration Board’s capability-building efforts led to a reduction in average processing times from 267 days to 85 days, saving more than $160 million annually.

Developing specialized capabilities in critical sectors

Governments’ investments in building expertise in particularly important or challenging vertical sectors, such as infrastructure or transportation, can have significant payoffs. Hong Kong’s Mass Transit Railway (MTR) developed deep expertise in core mass-transit capabilities such as operations, maintenance, and property management as a result of the government’s investment and rail-led transportation strategy. This expertise has helped the MTR, which is still 77 percent government-owned, to win contracts to maintain, operate, and improve metro systems in Australia, Sweden, and the United Kingdom. Another increasingly important subsector is cybersecurity: recognizing this, the US Department of Homeland Security is collaborating with universities including Carnegie Mellon University and the University of Maryland to train a pipeline of approximately 30,000 professionals in cybersecurity.

Sharpening strategic and risk-management skills

Governments face large, intractable challenges with many dimensions, multiple stakeholders, and far-reaching ripple effects. Some are perennial issues, like national security, whereas others may be external shocks—a natural disaster, perhaps, or a pandemic. Regardless of the nature or origin of the challenge, such problems often affect—and require coordinated responses from—multiple parts of government. Leaders must be equipped to anticipate, assess, and react to these complex problems. That was the impetus for the Singapore government’s creation of the Centre for Strategic Futures, which “aims to develop insights into future trends and discontinuities, and cultivate capacity and instincts to manage strategic surprises.”

In an increasingly complex and interconnected world, governments will struggle to address the challenges of doing more with less if their employees are not armed with the right skills. A commitment to capability building will allow governments to be able to take a more dynamic and adaptable approach to reform.

Closer collaboration with the private and social sectors

Finally, the public sector must adapt to a changing ecosystem in which the biggest challenges cross the boundaries of the public, private, and nonprofit sectors. The need for government to collaborate with the business and nonprofit worlds exists whether government is acting as a consumer of products and services, a provider of public goods, or an economic stakeholder.

Improving government’s procurement of products and services

Governments have come under fire for paying too much for products and services. Reducing the cost of government inputs (most of which are sourced from the private sector) is one of the main levers for doing more with less. One way the US government has accelerated the procurement process and lowered procurement costs is through its platform, which has been used by more than 50 federal agencies to invite companies and citizens to submit responses to “challenges”—in effect, requests for proposals. The top submissions, as judged by the requesting agency, are awarded a cash prize. The $10 million X Prize challenge supported by the Department of Energy yielded over $100 million in private investment. In response to the $50,000 Blue Button challenge, a company in just six weeks installed a function that made personal health records downloadable from a system used by approximately 200,000 doctors. With, the government gets more people thinking about how to solve tough problems, and it pays only for solutions that work.

Unleashing government’s power as a provider of public goods

Governments have for decades provided a set of public goods such as national defense and free-to-air TV. Open data—the release of massive, publicly-held data sets in machine-readable “liquid” form that can readily be used by developers—is arguably a new type of public good, with the potential to spur innovation among companies and other nongovernment entities. Todd Park, the US chief technology officer, has hosted “Datapaloozas”—events at which innovators and entrepreneurs build cost-saving apps using the more than 350 government data sets from 12 US health agencies, available on Similarly, opening up education data sets in several countries has spurred the creation of start-ups that use the data to improve teacher quality, reduce infrastructure costs, optimize school locations, and in general help educators do more with less.

Refining government’s role as an economic shaper and integrator

Governments have an opportunity—perhaps even a mandate, in certain troubled sectors—to play the part of a “systems integrator” that takes a high-level view on an issue and figures out how all stakeholders should work together. One area in which government can assume an integrator role is in the education-to-employment (E2E) system: our recent research on E2E has shown that 75 million young people are unemployed globally, yet only 43 percent of employers report that they can find enough qualified entry-level candidates. One integrator model is being tried in Brazil’s oil and gas industry. Prominp—a coalition of government agencies, private companies, trade associations, and labor unions—develops a rolling five-year projection of how much manpower is needed in specific geographies and skill areas (for example, shipyard welding or petroleum engineering), then identifies the best training provider to codevelop a curriculum with selected companies to meet those exact needs. By the end of 2012, Prominp had already qualified 90,000 trainees, and its goal is to bridge the projected skill gap of more than 200,000 skilled people by 2014.

The convergence of the public, private, and social sectors means that government leaders will increasingly need to be “tri-sector athletes,” adept in operating at the intersections of these sectors. And they will need to embrace new forms of organization and service delivery that are rooted in partnership.

In varying degrees, these four principles are already making a difference in local and national governments worldwide. But government by design isn’t easy. It requires political appetite and willingness to reform. It requires the readiness to try things that haven’t been tried before, and to quickly jettison ineffective ways of working. But the payoff—effective, affordable government that can better fulfill its multifaceted missions—will be more than worth the effort.

About the author(s)

Diana Farrell is the global leader of the McKinsey Center for Government. She is a director in McKinsey’s Washington, DC, office, where Andrew Goodman is a consultant.

More on Public Sector
Article - McKinsey Quarterly

Organizational health: A fast track to performance improvement

Article - McKinsey Quarterly

Culture for a digital age

Commentary - McKinsey Quarterly

Putting lifelong learning on the CEO agenda

Commentary - McKinsey Quarterly

Getting ready for the future of work