Skip to main content

From promise to delivery: Overcoming the strategy problem in the public sector

To deliver better strategies and outcomes, governments need to tackle social dynamics in the cabinet room.

The purpose of government is to improve the lives of citizens and make society better off for the future. That is true in every aspect of public-sector responsibility, from healthcare reforms that attempt to enhance patient well-being to efforts to reduce carbon emissions and initiatives to promote economic growth. As a result, government leaders often paint a “hockey stick” into the future; after an initial period of little change, sharp improvements in performance are forecast two or three years in the future.

Unfortunately, few governments are able to deliver the promised impact. Yet that does not deter public-sector planners from making equally bold forecasts for subsequent years. As a result, unrealized hockey sticks string together and the ugly cousin of the hockey stick appears, the “hairy back.” Exhibit 1 shows that this effect is not an isolated phenomenon. Hockey sticks are drawn repeatedly, for everything from global economic growth projections to country-level productivity growth to unemployment reduction to cost estimates of major infrastructure projects. The optimistic projections continue, even in the face of actual results that have stayed roughly flat—or worsened.

We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com

Why does this happen? The book Strategy Beyond the Hockey Stick shines a spotlight on the human biases and social games that result in hockey sticks and hairy backs in the private sector—and hold the majority of companies back from effective strategy development and execution.

The public-sector context is very different: at first glance it may seem that the lessons from private-sector strategy have little relevance for governments. Yet the dynamics in the cabinet room are not so different from those in the board room. Social games are perhaps even more deep-seated in the public sector than in private firms—and they often prevent governments from shaping and executing clear strategies to turn their forecasts into reality.

In this article, we draw on analysis by the McKinsey Center for Government (MCG) to highlight the challenges that most governments face in delivering on bold forecasts of improvement in economic or social outcomes. We also show that, as in the private sector, performance effectiveness is not equally distributed, with a small number of countries accounting for the lion’s share of high performance and improvement.

We then set out several common but poorly understood social dynamics present in public-sector cabinet rooms. We also propose several practical steps that governments can take to tackle those social games and refocus strategy on what really matters: better outcomes for citizens.

The unrealized opportunity in public-sector delivery

Governments the world over are striving to translate limited public-sector resources into better outcomes for citizens, in a context that is often complex and challenging. The results, unfortunately, are mixed. A look at the data on government performance at the country level shows there are dramatic differences in outcomes achieved—even among countries with similar development levels, political systems, and resource bases.

For the purposes of this article, we analyzed data on the healthcare and education sectors of 25 countries that represent 50 percent of global GDP and 12 percent of world population. Similar patterns exist in a range of other sectors, including transport, public safety, and tax administration.

In healthcare and primary and secondary education, we measured each country’s relative productivity. In Exhibit 2, we plot the relative performance of the 25 countries in healthcare for both 2005 and 2015. In our healthcare-productivity index, the country with the highest productivity scores 100 and the one with the lowest scores zero. In Exhibit 3, we plot a similar analysis for education productivity for 2009 and 2015.1

We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com

Consider the case of healthcare, where the distribution of productivity performance follows a clear curve. On the right-hand side of the curve are just a few countries—most notably the Czech Republic and Poland—that increased expenditure by less than $100 per capita for each year of life expectancy gained.2 On the left-hand side are three countries that increased expenditure by four times as much for the same gain in outcomes. The remaining 20 countries were stuck in the middle, achieving only mediocre healthcare productivity.

These findings point to a big opportunity: for governments around the world to replicate the achievements of the best-performing countries. As Exhibit 2 suggests, however, that is no easy feat. Very few countries have improved healthcare productivity over the past decade, and the shape of the curve remains—with a few countries significantly outperforming the rest.

In primary and secondary education, the picture is nearly as stark, as Exhibit 3 shows. The most productive countries—Australia, Canada, Estonia, and Poland—achieve excellent student outcomes at around half the spending per student of many other nations. The least productive countries spend triple the amount per student, yet achieve worse outcomes. Between 2009 and 2015, there was some movement in the middle three quintiles of countries, but the shape of the education productivity curve remains.

We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com

Strategy Beyond the Hockey Stick shows that a similar distribution of performance exists in the private sector: around 20 percent of companies account for 90 percent of economic profit. Perhaps that is to be expected: private-sector companies compete for market share, and only a few firms in any given industry are likely to achieve outsize profits. In the public sector, by contrast, such competition is limited to areas such as tourism market share and investment attraction. In core public-sector responsibilities such as health and education, outperformance by some countries does not come at the expense of others.

Indeed, best practices in public health and education systems can, theoretically at least, be replicated right across the globe. However, it is not easy: there was little improvement, and even some decline, in the educational productivity of most countries between 2009 and 2015. To achieve the promised delivery, a new approach is required across arenas of public-sector responsibility. First, though, they will have to tackle the social dynamics that cause hairy backs in the first place.

Social games in the cabinet room: The overlooked barrier to better outcomes

Why do so many hockey-stick forecasts of improved outcomes end up as hairy-back charts? In Strategy Beyond the Hockey Stick, the authors distilled the problem down to a big overlooked issue: the social side of strategy. Most leaders assume that their strategy processes are built on dispassionate analysis of data, trends, and opportunities, but social games played in the background are often the biggest driver of what gets decided and what gets done.

Our experience in working with governments around the world suggests that those social games and dynamics are just as present in the public sector as in the private sector. Let’s consider a few such dynamics described in Strategy Beyond the Hockey Stick and how they play out in government strategy processes.

The inside game

The inside view often prevails in cabinet rooms because they are tightly sealed. What comes into the room is basically what the participants bring in with them. That is generally a great deal of relevant experience, carried in the brains and memories of a few senior government leaders. A lot of data and information comes into the room, too, but it is typically focused on the participants’ own country and sector. A lot of information—including on relevant lessons from other countries and sectors—stays outside the room.

Nobel laureate Daniel Kahneman explained in his book Thinking, Fast and Slow how the realities of the outside world can disappear and be replaced by what he labeled “the inside view.” The inside view leads people to extrapolate from their own experiences and data, even when they are attempting something they’ve never done before. Kahneman says even he has fallen victim to the bias while designing a new syllabus and textbooks for the Ministry of Education in Israel. The team, relying on their experiences in other endeavors, initially projected that they would finish in one and a half to two and a half years. When Kahneman looked at how similar teams had performed on similar projects, he learned that 40 percent of teams never finished and that those that did needed seven to ten years.

Social biases

Strategy might seem as though it should be a primarily intellectual exercise. Yet strategy problems are exactly the low-frequency, high-uncertainty problems for which the human brain is least adapted.

People are prone to many well-documented unconscious cognitive biases—such as overconfidence, anchoring, loss aversion, confirmation bias, and attribution error. These biases exist to help us filter information for decision making. However, these unintentional mental shortcuts can also distort the outcomes, when we are forced to make big, consequential decisions, infrequently, and under high uncertainty. And these are exactly the ones we confront in the cabinet room.

Let’s consider a few such biases described in Strategy Beyond the Hockey Stick as well as examples of how they play out in government strategy processes:

  • Halo effect. “Our 3 percent increase in tax revenues last year reflected the progress of our modernization program and our decision to continue investing in digital.” A revenue service gives itself a pat on the back even though the whole economy also grew by 3 percent.
  • Anchoring. “We forecast 8 percent growth in tourism arrivals next year, plus or minus 1 percentage point, depending on the global demand environment. We will achieve this by pushing even harder on our current projects.” So 8 percent is the starting point of the negotiation, whether or not it should be.
  • Confirmation bias. “We’ve put lots of work into analyzing the reasons why this will work” [but no work into the reasons why it won’t]. “We have an electoral mandate for this plan” [so it can’t be argued with]. “We’ve also heard that country X is exploring this opportunity” [so it must be a good idea]. Good luck trying to stop the momentum for that project.
  • Champion bias. “Our inspiring leader has given us the audacity to dream. He/she has delivered on projects like this before—in their state/city/community. You should have the confidence in us to do it again at the national level.” This deflects attention from the merits of the project alone.
  • Loss aversion. “We know that governments that try and fail to deliver change are seldom forgiven—so let’s not risk anything too radical. The government needs to be a bedrock of stability and continuity, and ultimately we think the risks of bold new strategies outweigh the benefits.” Even though “business-as-usual” approaches might be putting the government at even greater risk.

Social dynamics

As hard as it might be to overcome those individual biases, they are only a part of the reason why you can’t just understand the social problems of strategy and assume that you’ll then be able to overcome them. Yes, as soon as you introduce people into strategy, you get biases. Then, when you introduce other people—that is, when the approver is different from the doer—you get agency problems.

Agency problems are fueled by incongruences between public-sector leaders and other stakeholders. Here are just a few of the more prominent ways that leaders and managers may act in their own interest, and not purely in that of the country, the public, and their organization:

  • “Sandbagging.” “I’m not going to put my neck on the line. I’m only going to agree to a plan that I know for sure I can deliver. My reputation is on the line, and I can’t risk being the one ministry/department that misses budget.” The reality is that individuals will often have a different attitude to risk than their overall organization does.
  • “The short versus the long game.” “Someone else will be running this ministry in three years, anyway. I just need to drive performance for the next couple of years and move onto a bigger role.” [The political appointee’s outlook.] “I’ll be working in the civil service for the next 20 years, long after this minister and this program have been forgotten—I have my career and my network to look after.” [The senior civil servant’s outlook.] The motivations of individual leaders are not automatically aligned to those of the government as a whole.
  • “My way or your problem.” “I know this department and the public sector better than the minister or the cabinet. They’ll just have to believe what I tell them. If I say it’s too hard, it’s too hard. If I don’t get the resources I ask for, then there’s my excuse for not delivering.” Line managers have inside knowledge, and often government leaders have little choice but to accept their version of the truth—and their advice.
  • “I am my numbers.” “I get judged by my numbers, not by how I spend my time. I’m just going to work hard enough to hit my targets, but not a lot more.” One’s minister can’t directly observe the quality of effort, and results can be noisy signals—were those poor results a noble failure; were those great results dumb luck? Civil servants often maneuver to ensure targets are easily achievable.

While governments hope their people are all pulling in the same direction, in reality they have very different motivations and certainly asymmetric information.

Let’s not forget incentives, either. These go way beyond financial remuneration. Presenting in front of your superiors or your peers is a matter of pride. Your track record is a matter of ego. Your team wants protection. As Warren Buffett has said: “95 percent of behavior is driven by personal or collective incentives.”

Strategy involves a complex set of motivations in a complex game. Far from having a single goal that everyone can focus on, government leaders are negotiating next year’s budget, competing for resources, delegating responsibilities to others, maintaining and escalating prior commitments, impressing the public and the media, inspiring confidence among a broader set of stakeholders—all at the same time. They know that they have to craft a strategy that claims to generate a 15 percent improvement to get the 10 percent they really want, and they know that the main act is the budget. The strategy discussion is just the opening salvo.

One of the most widely read pieces of research that McKinsey has published in the past decade showed that companies that rapidly reallocate capital to new growth businesses outperform those that take a steady-state approach. Yet, the social side of strategy is such that companies still tend to take what is known as a “peanut butter” approach—spreading a thin layer of resources smoothly across the whole enterprise, even though it’s clear that opportunities are far greater in some areas than in others. We observe similar behavior in the public sector, where it is often hard for high-potential new initiatives or innovations to secure the funding they need to scale—because there is great pressure to leave the budgets of existing programs untouched.

No matter the precise motivation, government leaders and managers will use every bit of social power they can to improve the chances of their ministry, department, or unit succeeding. Even if we don’t like to acknowledge it, we are social creatures and covet status in the tribe. This was an excellent trait from an evolutionary perspective, but can be an obstacle when developing good strategies.

How governments can tackle social games and get to better strategies and outcomes

However daunting the social side of strategy might seem, our research and experience suggests five specific shifts that can dramatically improve the quality of strategic choices made in the cabinet room. Individually, they are straightforward enough to be implemented tomorrow. Taken together, they enable government leaders to change what is happening in the cabinet room—and avoid another hairy-back curve.

1. From ‘getting to “yes”’ to ‘debating real alternatives’

Most strategy discussions bring one plan into the room. The goal is either to approve or reject this single proposal. Diverging opinions, or questions about the underlying assumptions, are often unwelcome in this conversation. Yet, in taking this approach, confirmation bias and groupthink are inevitable: all focus is on one choice set. That directly stands in the way of the deeper reflection that is needed to have difficult conversations and make hard choices.

Organizations therefore need to systematically use debiasing techniques to stress test their strategies—and spark real debate. For starters, public-sector leaders could use classic debiasing techniques. For instance, they could ask different subteams to work out different plans (focusing on the most difficult choices). Or, public-sector leaders could ask the team to describe all the different assumptions that need to be true to succeed (and then develop alternative views from this).

Public-sector leaders could also cast a wider net, for instance, by involving the public directly. A good example of this featured in our government productivity report is the MyGov online platform launched in India in 2014. The platform invites citizens to share comments, ideas, or concerns. To date, nearly 7.9 million citizens have participated, submitting suggestions in policy areas ranging from environmental pollution to girls’ education to health. One proposal submitted through the platform was to turn rural post offices into simple banks to boost financial inclusion, and it was included in India’s 2015 budget. By March 2017, banking sections had been installed in 25,000 post offices. Such participative planning puts citizens at the heart of designing effective outcomes—and thereby opens up the debate.

2. From ‘try to please everyone’ to ‘have the courage to prioritize’

In our article “Delivering for citizens: How to triple the success rate of government transformations,” where we interviewed almost 3,000 public officials, we emphasized that governments exist to serve all their constituents equally. As a result, their responsibilities span a great many areas, from education to security to transport to public finances. Unlike private companies, they cannot choose to focus on “core business” or to exit sectors or geographies that are challenging. This broad mandate puts pressure on government leaders to try to please everyone—setting more objectives than it is possible to deliver, and launching more programs than can reasonably be completed.

Yet, many of the successful government leaders we have spoken to have emphasized the power of prioritization. One is Prime Minister Ana Brnabić of Serbia, who clearly focused on two core priorities when coming into office: digitization of government and transformation of education. Similarly, Rachna Gandhi, executive of an Australian government agency, told us her institution “would never have got up if we had taken the approach of, ‘let’s get everything right.’”

As our research showed, when countries manage to prioritize the effect can be significant. For instance, a regional government in an African country sought to accelerate economic growth and create jobs on a large scale. It identified more than a dozen sectors that had promise, but realized that, given limited resources and leadership bandwidth, it would need to focus on only a few if it were to move the needle on growth and jobs. The prioritization was fraught, with several stakeholders arguing that government had no right to “pick winners.” Yet the narrow focus paid off. In one of the selected sectors, tourism, the region achieved record visitor numbers within two years—thanks in part to government interventions such as facilitating more international flight arrivals, ramping up destination marketing, and expanding the region’s capacity to host conventions.

3. From ‘budget inertia’ to ‘liquid resources’

Many organizations rule by last year’s budget, with only slight changes. As a result, a budget is not able to adapt to a need of a dynamic environment. MCG’s research shows the public sector is particularly prone to this tendency, often simply rolling over last year’s budget. From 2006 to 2014, for example, 92 percent of the annual sector-level budget allocations of EU countries, adjusted for inflation, changed by less than 1 percent from one year to the next (as highlighted in Exhibit 4).

We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com

Without breaching public-sector rules that govern how budgets can be reallocated, we believe there is room for much greater flexibility in budgeting and adjusting resources to the true needs of the priority or program as it is implemented.

For example, as we discussed in our government productivity report, Singapore’s Ministry of Finance produces the biennial Singapore Public Sector Outcomes Review, which reports on whole-of-government outcomes and indicators that reflect policy priorities such as monthly household income and work-training participation rates. These indicators, in turn, inform budgetary decisions in future years.

Australia allows departments’ unspent administrative costs to be rolled over to the following year, subject to cabinet approval. Other countries allow departments to accumulate carryovers, subject to prescribed limits. France and Sweden, for example, allow a maximum of 3 percent of each department’s expenditure to be carried over from one year to the next.

Another way to infuse greater flexibility or agility into budgeting is to earmark a small portion of the budget of a department—say 2 to 5 percent—and allocate parts of it on a quarterly basis to addressing emerging priorities or for special projects based on ideas from civil servants at the front line.

4. From ‘sandbagging’ to a ‘holistic performance management’

When people develop strategic plans, they often skew the target to something that is firmly within range. Because of this risk avoidance, plans often fail to prioritize higher-risk, higher-reward efforts. The consequence is that promising reforms or cutting-edge innovations are often thwarted, as no one in the organization is willing to take the risk. In the public sector this is a particular challenge. Often the rewards scheme is limited (creating little upside for success), or the required performance management itself is lacking altogether.

The moment members of the organization are willing to give up sandbagging, they need to be rewarded for it. Unfortunately, incentive schemes often forget to incorporate the relative probability of success in its equation. For instance, meeting the target of a “no-regret move” oftentimes gets similar appreciation as achieving a stretch target. Bringing probabilities to the fore can help change the dialogue in public-sector strategy setting.

It is also important to move toward holistic performance management, which encourages noble failures and focuses on the quality of effort. As we noted in our government productivity report, Singapore assesses its civil servants using criteria adapted from high-performing private companies as well as leading public-sector organizations such as the US Air Force. The process produces both a backward-looking “performance grade” and a forward-looking “potential score.” Both are benchmarked against each person’s own performance and those of his or her peers.

5. From ‘long-range planning’ to ‘forcing the first step’

A “perfect plan” does not exist. Indeed, the pursuit of one will likely cause paralysis: when plans and targets become too rigid and mechanistic they can become an obstacle to successful transformation. For many public-sector leaders, planning is a particularly difficult endeavor, given the multiplicity of stakeholders. In a January 2018 interview with McKinsey, then Deputy Prime Minister of Slovenia Boris Koprivnikar put it this way, “Transformation is a bit like climbing a mountain. If you want to climb a mountain, first pick the mountain and then start walking. Don’t plan too much because you don’t know what kind of roads and obstacles and opportunities you will get. You need to know where you are going, what is your final goal.”

With that in mind, we recommend that planning is treated like multiple sprints— rapid, time-bound processes of just a couple of months—with clear deliverables and a non-negotiable deadline. Our experience with public-sector clients shows that officials should not spend months in planning working groups; plans can be changed and improved once implementation begins. Delivery, on the other hand, should be seen as a group marathon, with a steady pace and momentum sustained over time.


Working in government provides an opportunity to contribute to better outcomes for citizens and society—and drawing on that profoundly positive motivation should be at the heart of any public-sector strategy process. Yet social games often cloud that sense of purpose and sap public servants’ motivation. Understanding and tackling those games is a critical step in unlocking governments’ full potential to help foster healthier, more prosperous, and more sustainable societies.

We are convinced that the shifts we discuss here will free governments to define and deliver truly effective “big moves,” including initiatives that drive transformational change, reallocation of resources to the most important priorities, and unlocking public-sector innovation. In the next article in our series on strategy in the public sector, we shine the spotlight on the power of big moves and celebrate the pioneering governments around the world that are already making such moves.

About the author(s)

Daniel Cramer is a consultant in McKinsey’s Amsterdam office, where Johanna Hirscher is a partner and Sven Smit is a senior partner. Gundbert Scherf is a partner in the Berlin office.

Related Articles