 Earlier this week, President Obama announced that he was committing $70 billion of government money to create a new manufacturing innovation hub in the city of Chicago. About the same amount will be given to the city of Detroit, and these two will join existing government-funded manufacturing hubs in Youngstown, Ohio and Raleigh, North Carolina. These initiatives are part of a larger program proposed by the president last year to devote about a billion dollars of federal money to the creation of a series of high-tech manufacturing clusters in the United States, and the idea is that these would also bring in private investment to complement the government's dollars. Essentially, what the president wants is to create a bunch of Silicon Valleys in the manufacturing and technology sectors in the US. Everybody loves Silicon Valley. Every government planner hopes to have something like Silicon Valley in their own territories. Why? Because innovative clusters like Silicon Valley have a number of benefits for the firms and other institutions that locate there. There are many benefits from what economists call co-location, the exploitation of spillovers among firms and labor markets that are involved in similar activities. That's why lots of tech firms want to be in places like the San Francisco Bay Area or Seattle or Boston's Route 28 or Austin Texas, other areas where they have a lot of similar firms that are located nearby. This allows firms to share specialized workforces to benefit from all kinds of tacit interactions and they link to benefit more generally from having a lot of like-minded folks nearby. The problem is that technology clusters, like other kinds of clusters, tend to emerge from the bottom up. They cannot be designed and planned from the top down. No one knows exactly where an innovative cluster will form. This is something that only the market and experimentation can decide. In 2008, the president allocated $249 million federal stimulus money to an electric battery company called A123 Systems to make batteries for electric cars. Four years later, that company, along with other electric battery firms that had received federal funding, declared bankruptcy. The government has no way of knowing what kinds of activities and what kinds of firms will be successful in the competition for resources in the free market. Government planners don't have skin in the game. They're not investing their own money in anticipation of the development of a cluster that will generate high returns. They're investing your money. They don't have skin in the game and there's no reason to think that government planners would make better decisions than private investors and other ordinary market participants in deciding what infrastructure should be in place, where companies should locate, and how companies should try to benefit from locating close to each other. What can the government do to try to stimulate innovation, to try to make the U.S. or whatever economy more competitive? The best that government can do is get out of the way. The government should avoid placing obstacles in the paths of investors and entrepreneurs who want to create technology clusters. That means don't create business cycles by inflating the currency, expanding the money supply, and leading to malinvestments in particular places and industries. Don't bail out failing industries. Don't stymie the creation of new businesses through heavy taxes and burden some regulation. Rather, the government should step aside and let the private sector do what it does best, namely creating economic value for consumers.