CRS Gets It Wrong Again on Manufacturing: Not All Industries Are Created Equal
ITIF has a long tradition of working on manufacturing policy. Our U.S.-focused work has been in two main areas: The first is to show how U.S. government statistics on manufacturing output are flawed, which leads to an overstatement of output growth and causes many analysts to conclude incorrectly that trade has had almost no effect on the decline in manufacturing employment in recent years. The second is to promote an advanced manufacturing policy agenda to support the next generation of manufacturing in the United States. This includes programs such as the National Network of Manufacturing Innovation, the Manufacturing Universities program, and the National Institute of Standards and Technology’s Manufacturing Extension Partnership (MEP).
We have done this work in part to challenge the philosophy that so many Washington economists subscribe to: “car production, car rental, what’s the difference?” (Or, as President George H.W. Bush’s economic advisor Michael Boskin put it, “computer chips, potato chips, what’s the difference?”). According to this view, all industries are alike—they sell things and they employ people, and manufacturing is no different. If policymakers buy into this dubious notion, then why would they support manufacturing policies? As economist Martin Feldstein once asked me in the 1990s when we debated the merits of the United States having a program like MEP to help U.S. manufacturers compete globally: Why not have one for restaurants, too? My answer, which didn’t seem to convince him, was that we don’t face competition from Japanese restaurants.
The latest entry in these “all industries are the same” chronicles comes from the Congressional Research Service (CRS) in a report authored by Marc Levinson. We have critiqued Mr. Levinson’s prior work attempting to justify massive U.S. manufacturing job loss as normal. Now he has authored a new CRS report that advises policymakers to ignore manufacturing policy because it doesn’t create many jobs. But as with the prior CRS report, there is much to dispute here.
First, the CRS report engages in circular logic. Manufacturing doesn’t create that many jobs, because U.S. manufacturing competitiveness lags many other nations, which is partly because of an unsupportive policy environment, including a very high corporate tax rate, a weak R&D tax credit, limited manufacturing tech and skills programs, and more. The report’s logic is that since manufacturing is not creating that many new jobs, we shouldn’t have policies for it. But the reason it is not creating as many jobs as it could be is because the United States doesn’t have the right policies.
Second, the CRS report completely ignores the economic concept of multiplier effects. This is made especially clear when it says that manufacturing is not all that helpful for local economic development because there are fewer large factories than there used to be. First, a key reason we have fewer large factories is lack of U.S. competitiveness. Second, the economic research is clear that because manufacturing is an export-based industry (most sales are outside the local labor market), it supports jobs in local services industries (e.g., barber shops, grocery stores, doctor’s offices, and more). This is economic development 101. You see this dynamic play out at the state level, too. That is why I wrote that it’s time for Washington to think like a state:
Because successful traded-sector firms play an outsized role in a state’s economic fortunes, losing one, whether because it goes out of business or because it moves, also has an outsized impact. Not only does that firm lay off its own workers, but also its absence lowers demand for products and services of other local firms (e.g., banks, janitorial service providers, printers, etc.), and both the workers it leaves behind and the workers at these other suppliers all have less money to spend on everything from haircuts to restaurant meals. This “multiplier effect” can be very large as it ripples through a state economy, causing a downward spiral of economic contraction and job loss. That’s why all states have policies that are targeted to help firms in traded sectors such as manufacturing, tourism, software and Internet-based services. In Washington, however, most policymakers and most of the economists who advise them do not think this way.
Finally, as with previous reports we have critiqued, this new CRS report refuses to engage with the scholarly literature on how U.S. government statistics overstate manufacturing output. So, when it says that manufacturing output is up significantly in the last decade but jobs are not, it implies that virtually all output growth in U.S. manufacturing is a reflection of higher productivity, not growth of employment hours. In fact, this is an overstatement. If we could grow U.S. manufacturing output in a real, not statistically flawed way, then the United States would enjoy more jobs.
The report claims that U.S. policymakers focus on manufacturing because of its employment potential. And then says this is a fool’s errand. To be sure, even if the United States had a robust competitiveness and manufacturing agenda and eliminated the manufacturing trade deficit, it wouldn’t create massive numbers of jobs. But it would create some. In fact, we have estimated that if the United States could eliminate the trade deficit in manufacturing, even by raising productivity 20 percent, it would create 1.3 million direct jobs and many millions more indirectly through the employment multiplier. But the more important point is that the reason for a national manufacturing strategy should not be to create jobs, even though such a strategy would produce considerably more jobs than Levinson would have us believe. The reason for a strategy is to boost U.S. competitiveness and eliminate the trade deficit, which is a debt that future generations of Americans will have to pay otherwise. For all of these reasons, we need a robust national manufacturing strategy.
