Christian Ketels, Harvard Business School researcher

Economic development should spur region-wide productivity, not special deals for some

A Civic Caucus Focus on Competitiveness Interview

September 27, 2013


John Adams, Audrey Clay, Janis Clay, Paul Gilje (coordinator), Dan Loritz (chair), Dana Schroeder, Clarence Shallbetter. By phone: Richard Bryden, Christian Ketels, Tim McDonald, Marcela Merino.


According to Harvard Business School researcher Christian Ketels, the old view of economic development was that government's role was simply to attract new businesses. But, he says, the new view is that economic development is the process of improving the business environment to enable companies to compete in increasingly sophisticated ways. New trends, which are changing the reality of state economic development, call for strengthening the role of regions in fostering economic development; emphasizing job creation from within; supporting advanced manufacturing; helping meet industry demands for talent; raising expectations for universities to foster commercialization from research; and increasing business export initiatives.

He reports that Minnesota is one of 15 states with high and rising prosperity compared with the U.S. But Minnesota should work on its labor force productivity, which is below the national average. In recent years, Minnesota has retained its high rankings among the states on four of eight economic measures, but the state's rankings on seven of the eight measures are trending downward. He notes that, despite ranking low in overall cluster strength, Minnesota had five business clusters that ranked high nationally in 2009: medical devices, publishing and printing, analytical instruments, processed food and information technology.

Ketels emphasizes that regional economic development must ensure more competition and rivalry within and between business sectors. He says less rivalry, which may make an individual company more profitable, is not a good thing for a region, because it reduces the pressure for higher productivity and more innovation. He argues that government's role must be to create an environment where businesses can be more productive and compete more efficiently. Giving special deals or free access to government money to certain businesses does not increase productivity.


Christian Ketels is a member of the Harvard Business School faculty at Prof.Michael Porter's Institute for Strategy and Competitiveness. He is president of TCI, a global network of professionals in the field of competitiveness, clusters and innovation. He is Honorary Professor at the European Business School Oestrich-Winckel and Senior Research Fellow at the Stockholm School of Economics.

Ketels has led cluster and competitiveness projects in many parts of the world, has written widely on economic policy issues and is a frequent speaker on competitiveness and strategy in Europe, North America and Asia. He holds a Ph.D. in economics from the London School of Economics and further degrees from the Kiel Institute for World Economics and Cologne University.


Translating analysis of regional economies to real change is the hard part. "We have good tools for analyzing regional economies," began Christian Ketels, "but translating analysis to real change is still somewhat of an art form. It depends a lot on the local context."

The old view of state economic development was attracting big investment. This focused on taxes, sites, infrastructure, skills, amenities and a creative class, Ketels said. "This is still the reality of what economic development is about in parts of the country."

New trends are changing the reality of state economic development. "We see that things are changing," he said. "There's more of a recognition that a lot of the work needs to happen within your region. It's great to attract investment and that's part of what you need to do, but a lot of the job creation comes from existing companies."

The new trends include:

  • Strengthening the relationship between the state and its regions in fostering economic development;
  • Emphasizing job creation within state borders;
  • Strengthening state support for advanced manufacturing;
  • Creating partnerships to meet industry demands for talent;
  • Raising expectations for universities to bridge the gap between research and commercialization; and
  • Increasing business export initiatives.

Even within national economies like the U.S., there's a huge variety of performance among economic regions that isn't easily explained. A comparison among states based on Gross Domestic Product (GDP) per capita and growth in GDP per capita from 2000 to 2012 shows rich states that are doing well and not so well and poor states that are doing well and not so well.

Minnesota is one of 15 states with high and rising prosperity compared with the U.S. Included in the same category are states like Massachusetts, New York, Texas and North Dakota. "You really need to understand not only national policies, but also what's going on in the states," Ketels said.

Local factors really matter. He said the same GDP comparison among economic regions within Minnesota would show a similar variety of performance.

Ketels explained that important local factors include geographic location; natural assets; size; density; culture/social capital; business environment quality, including infrastructure and skills; government capacity; and clusters.

He said research shows that key drivers of regional performance include:

  • Social capital. The culture and the trust you have in your society that enable you to draw on other assets are very important. This social capital seems to drive the economic performance of regions.
  • Skills. Performance is often better when you have access to a more educated workforce that meets the demands of your companies.
  • Specialization patterns. You need to think about your performance in areas in which you are strong. Enhance your capacity in those types of activities in which you happen to be strong and in which you happen to have strong specialization patterns.

Recent evidence shows the presence of clusters is related to strong economic outcomes. Ketels defined a cluster as a geographically concentrated group of interconnected companies and associated institutions in a particular field of related industries.

He said that there is now ample evidence that having a higher share of regional employment in strong clusters translates into:

  • Higher prosperity: Higher wages, higher productivity, higher job growth and more innovation.
  • Greater entrepreneurship: More new business formation in new and existing industries; better survival of new firms; and higher job growth in new firms.
  • Easier emergence of new clusters: Driven by the numbers and kinds of existing clusters.

There is no one silver bullet in achieving sustained economic growth. Ketels pointed out that a range of factors matters and that the individual factors are interdependent in their impact on regional prosperity. For example, he said, it's good if your region gets a research institution, but companies will only make changes in their activities based on that if you also have an advanced workforce. And it is those changes in companies that ultimately drive prosperity gains. He added that the value of education or infrastructure depends on how good the region is in other areas.

To move your region to action, you have to diagnose where the region stands economically. Ketels believes the diagnostic process needs to be open and inclusive of all relevant stakeholders. "You must be clear that this is a team effort," he said. It's important that the parties speak the same language and establish a common set of facts, that is, to agree on the baseline of where the region stands now economically. "Diagnostics are critical to make sure everybody gets to the same baseline," he said. "The analytics need to be sound, but having an expert who is 100 percent right is of little value if there is no ownership of the findings among regional decision makers.

"You need to establish a common understanding on where the region stands. Then you can fruitfully debate the conclusions to draw," Ketels said, "and argue about what is right for this region now. If you don't agree on what the current circumstances are, that discussion will be all about ideology and individual beliefs, not about facts."

He outlined three areas the diagnostic process needs to cover:

  • Economic outcomes: How are we doing in exports, innovation rates, investment rates and other measures?
  • Competitiveness fundamentals: These are things that you can affect and that ultimately drive economics outcomes. Where do we stand on education, infrastructure, business environment quality, and quality of institutions?
  • Natural conditions: These are very important. "Part of your performance reflects the deck that you've been dealt: where you're located, who your neighbors are, whether you're on a trade route and what your natural resources are," Ketels said.

A 2012 scorecard shows that in recent years, compared with other states, Minnesota has retained high rankings on four of eight economic performance measures, but seven of these eight measures are trending downward.

Retained high rankings:

  • Labor Mobilization (Proportion of working age population in the workforce): Ranked 1st in 2000 and 2 nd in 2010.
  • Innovation (Patents per employee): 6th in 2000 and 6 th in 2010.
  • Wages (Average private wage): 11th in 1998 and 13 th in 2009.
  • Prosperity: (GDP per capita): 13th in 2000 and 12th in 2010. This is the only measure whose ranking improved in recent years.

Currently at middle rankings:

  • New Business Formation (Traded cluster establishment growth): 15th from 1998 to 2000 and 19th from 2007 to 2009.
  • Labor Productivity (GDP per workforce participant): 19th in 2000 and 21 st in 2010.
  • Job Creation (Private employment growth): 20 th from 1998 to 2000 and 24th from 2007 to 2009.

Retained low ranking:

  • Cluster Strength (Employment in strong clusters): 30th in 1998 and 39 th in 2009.

Despite ranking low in overall cluster strength, Minnesota had five high-ranking clusters nationally in 2009 (measured by employment size): medical devices (4th); publishing and printing (7th); analytical instruments (8th); processed food (9th); and information technology (12th).

Minnesota should work on its labor force productivity. Looking at productivity as measured by national GDP per labor force participant, Minnesota ranks below the national average. But its productivity grew slightly faster than the national growth rate from 2000 and 2010. "This should be on Minnesota's action agenda," Ketels said. "Why is our productivity not higher and how can we drive this forward?"

Minnesota ranks higher than the national average in innovation performance (measured as patents per 10,000 workers) and its performance is improving at a faster rate than the national average. "This is so important," Ketels said, "because this is where policy can act." There are other performance measures that a region can't address directly.

Economic development is the process of improving the business environment to enable companies to compete in increasingly sophisticated ways. Referring to the "diamond" of business environment conditions, a framework introduced by Michael Porter in the 1990s, Ketels noted that the quality of the overall business environment can be addressed directly through:

  • Rules and incentives that encourage local competition, investment and productivity, such as tax policy that encourages investment and research and development (R&D), flexible labor policies, intellectual property protection and antitrust enforcement.
  • Access to high-quality business inputs, such as human resources, capital access, physical infrastructure, permitting and regulatory efficiency, and scientific and technological infrastructure.
  • Local availability of suppliers and supporting industries.
  • Addressing sophisticated and demanding local needs and customers through things like strict quality, safety and environmental standards; consumer protection laws; government procurement of advanced technology; and encouraging early demand for products and services.

Ketels said that economic development attempts must address the fundamental conditions of the economy at this level to have a sustained impact. Short-term measures to artificially create jobs or support exports do not work over time, if they do not change the underlying competitiveness of the location.

Regional economies are characterized by two kinds of clusters: traded clusters and local clusters. Both are important, Ketels said, but they have very different dynamics and react very differently to changes in the business environment.

  • Traded clusters are companies that serve national and international markets. They can locate anywhere and represent 30 percent of employment. "If they don't find Minnesota a good place to locate activities, they will go away," Ketels said. "They represent only 30 percent of employment, but it's the higher wage part, the much higher productivity growth part, the much higher innovation part. More than 95 percent of patents are in these sectors."
  • Local clusters are companies that serve almost exclusively the local market and are not directly exposed to cross-regional competition. They represent 70 percent of employment.

Government action defined the old model of economic development; collaborative action defines the new model. Under the old model, Ketels said, government was seen as the driver of economic development through policy decisions and incentives. Under the new model, economic development is a collaborative process involving government at multiple levels, companies, teaching and research institutions and private sector organizations.

"There are many actors who have to be involved and who have relevant information to identify what needs to be done in a specific location," Ketels said. "It's a collaborative process where you have to have trust, a shared language and a common view of reality. Because there are no generic 'silver bullets' for economic development, you need to start with the diagnostics of where you stand as a location. To then get to a joint agenda, the collaborative process is already critical at that stage."

"But, Ketels continued, "it's not only the diagnostics and policy design, it's also the approach towards action that needs to change: in this new model there is a recognition that many individuals need to act to make change happen and that you can't order them to do what is required. You can only try to convince them. The hope is that people will do things trusting that other people in the system will do complementary things. That will make it more valuable for me to make that investment."

Minnesota contains parts of six economic regions, all of them spilling into or out of neighboring states. The six regions are (1) the Minneapolis-St. Paul economic area, which spills into Wisconsin and South Dakota; (2) the Duluth economic area, which spills into Wisconsin; (3) the Grand Forks economic area, which spills into Minnesota; (4) the Fargo economic area, which spills into Minnesota; (5) the Sioux Falls economic area, which spills into Minnesota; and (6) the La Crosse economic area, which spills into Minnesota.

Different levels of government have different roles in economic development. "Do you have a clear view of what the regional/local level can and should do?" Ketels asked. "The regional/local level leverages the investments and tools provided by the national and state governments. Should we try to attract funding for a new community college? Should we try to get money for a new research center? Or should we go after some new infrastructure investment? You need to decide at the local/regional level what you want to try to get from higher levels of government."

Regional government can use cluster data and cluster groups to decide what dimensions of the business environment to strengthen, what networks to mobilize and what skills, firms and investors to attract. Ketels said cluster-based policy does much more to put the region in focus, build on the region's existing capabilities and move policy design and implementation into a public/private partnership, where the different actors work together.

The state government has more tools, power and capacity to do economic development, but, in terms of businesses and clusters, the state is not the most relevant unit. Ketels asked whether the state can do things that work in different parts of the state or whether it should give different parts of the state the ability to do things themselves. If it stays at the state level, there must be a strategy that acknowledges differences among different areas of the state. If the state gives substate regions the power to do things themselves, there must be capacity at that level to do these things. "Otherwise," he said, "you're pushing things down into a void."

There must be competition within sectors and between sectors. An interviewer asked about how to account for breaks for certain companies, import restrictions, public pensions and other things that tip the balance away from a competitive environment to a noncompetitive one. Ketels said we must distinguish between tools that make economic activity in businesses that form a sector privately more profitable vs. those that make activities in those sectors more productive. The sectors argue for a lot of the subsidies and market restrictions because it makes their activities more profitable. But it doesn't make them more productive.

"We must be very clear that we're not primarily interested in the profitability of your company; that is your job as a business leader, " he said. "But we want to create an environment where you can be more productive. That doesn't mean special deals or free access to government funds."

"It's important that government leaders understand the principle of competitiveness," he continued. "Competitiveness doesn't mean special deals to the sectors that are most important to our economy, but it means making everyone more productive and more able to compete effectively."

He said less rivalry may seem good individually, because it makes an industry more profitable for an individual company. "But it's not a good thing for a region, because it takes away from the pressure for higher productivity and for higher levels of innovation. If we come together for regional economic development, we want to make sure there is more competition and rivalry, not less."

Incentives to businesses from local or state governments basically shift money from one to the other, because the governments think there will be more benefits than costs. Ketels replied that it's important to think about how to tie the investor to the location. It's better to provide more things to investors on the ground that would stay, even if an investor leaves. "If you think more about the regional economic strategy, we should try to make sure we're able to tell companies we're particularly good at certain things for their industry," he said. "Take it away from the kind of competition that's only decided by the lowest tax rate. You need to make sure there's a broader spectrum than that."

Because political units are not necessarily aligned with the real economic units, coming up with the right kind of institutions across these political borders is very important. Ketels gave the example of the decision to create a joint port authority between New York City and New Jersey. "It was absolutely critical," he said. "It allowed all kinds of things to move to New Jersey that created space for new things to happen in Manhattan. That wouldn't have been possible if there were still a New York City port authority that tried to defend its turf."

If things are happening that make Minnesota less attractive for certain companies, giving them money doesn't necessarily help in the long run. An interviewer pointed out that in the 2013 session, the Legislature gave the Mall of America, the Mayo Clinic and 3M major amounts of money, because these institutions are very important to us. Ketels said he is somewhat skeptical of efforts like that, but it depends on what the money is used for. "Ultimately you do want to create an environment where these companies want to stay because they think this is the best place to be," he said. "But giving them money can at best be a short-term part of a long-term strategy."

Comment here on this interview with Christian Ketels