The US is the second-largest recipient of FDI in the world, behind China, and by far the largest target for FDI among OECD countries (OECD 2013). The numbers are large ($253 billion for the US), and the gap with the next-largest in the OECD is impressive ($63 billion for the UK and $62 billion for France in 2012).
How do these vast inflows affect US workers and the US economy as a whole? This column presents evidence on the impact of inward FDI on the US domestic economy – in particular, on the quality of jobs in the US, on productivity and value-added by foreign firms in the US, on R&D undertaken in the US, and on spillover benefits to unrelated US firms. It is based on our recent study for the Peterson Institute for International Economics (Moran and Oldenski 2013 ).
A growing new phenomenon is inward FDI to the US from emerging markets, such as China, India, Russia, and Brazil. We also ask whether there might even be positive spillovers from emerging-market investors to US domestic companies.
Comparative descriptive statistics
In Moran and Oldenski (2013), we uncover a number of facts about the contribution of foreign-owned firms to the US economy. For example, foreign investors in the US pay higher wages than US employers. They even pay higher wages than US multinationals, which are among the highest-paying of all US firms. The average worker at a US firm earned $64,552 in wages and benefits in 2009. Compensation is slightly higher for the US-based employees of US-owned multinationals. They earned an average of $69,208 that year. However, the US-based employees of foreign-owned multinationals out-earned both of these groups, with an average income of $77,597. By paying higher-than-average wages in the aggregate – and usually higher wages than paid at US companies in the same industry – foreign investors in the US are creators of ‘good jobs’.
Turning to R&D, the data show that US firms spend about 4.8% of value added on R&D, whereas the majority-owned affiliates of foreign investors spend more than 7% of their US value added on R&D. This might seem surprising, since multinationals could be expected to concentrate their R&D activities close to their worldwide headquarters. Indeed, the US headquarters of US multinationals spend slightly more on R&D than do affiliates of foreign multinationals operating in the US. However, the fact that foreign firms spend so much on R&D in the US suggests that they are not overwhelmingly motivated to keep high-value-added activities near their home headquarters. International companies want to locate production activities where they will be performed the most effectively. The US’s abundance of highly educated workers and friendly environment for research and innovation make it a desirable place to locate those activities – whatever the nationality of the parent.
Spillovers from foreign investors to domestic firms in the US
In addition to the direct benefits, such as paying higher wages and investing in R&D, foreign firms in the US may have more indirect effects on the US economy as well. For example, these firms may bring new production techniques and management know-how that could ‘spill over’ to domestic firms, making these local companies more productive as a result.
In Moran and Oldenski (2013), we provide new empirical evidence of these spillover effects. We use a methodology proposed by Keller and Yeaple (2009), but with more recent data and a focus on the level of economic development of the FDI source country.
We use firm-level data on characteristics of domestic firms from Compustat, and data on FDI from the US Bureau of Economic Analysis. The dependent variable in this analysis is firm-level total factor productivity (TFP) of US firms. If foreign firms are inducing greater efficiency in their domestic counterparts – either by exposing them to new technology and business practices, by demanding higher-quality inputs, or by increasing competition – these effects should show up as changes in TFP, which essentially captures how efficiently firms use a given set of inputs.
Conceptually, this measure of TFP is simply the difference between the actual output of a firm and the output that would be expected given that firm’s capital, labour, and material inputs, where the expected effect of each input is calculated controlling for the endogeneity of productivity and input choice, as well as the possibility of firm exit following the methodology of Olley and Pakes (1996).
But industries still differ in important ways – such as their use of technology or the ease with which their production can be fragmented across borders – that might affect both productivity growth and FDI. Omitting these characteristics could potentially make it appear as though FDI were causing the productivity growth, when in reality investment and productivity changes were both due to some omitted third variable. For this reason, industry-level fixed effects are included to hold constant the identity of the industry – and thus all of its defining characteristics – to ensure that these characteristics are not confounding the analysis. Similarly, year fixed effects are included to control for macroeconomic factors and other characteristics that will vary over time in the sample.
The results provide strong evidence of positive spillovers from direct investment in the US. An increase in the share of US employment accounted for by foreign firms in a given industry is positively and significantly associated with growth in total factor productivity. In other words, the presence of foreign firms makes domestic US firms more productive. This result holds for FDI lagged by both one and two years, but the magnitude of the effect is greater after two years, suggesting that the benefits of positive spillovers are increasing over time. We find that a one percentage point increase in the share of total employees in an industry who work at foreign-owned firms in the US increases the productivity of all firms in the industry by an average of 0.81% after one year and by 2.75% in the second year, or a total of more than 3.5%.
To put the magnitude of the spillover effects in context, consider the total impact that inward FDI has had on US productivity growth over the past two decades. Overall, US total factor productivity grew by about 25% from 1987 to 2007 (OECD 2011). Over that same time period, employment at foreign-owned firms as a share of total US employment grew from about 3.8% to 4.6% (a 0.8 percentage point increase). Using our estimates, that implies that productivity spillovers from FDI alone were responsible for US TFP growth of about 3% (0.008*(0.81+2.75)) from 1987 to 2007. This 3% is more than one-tenth of the 25% US TFP growth over that period. In other words, about 12% of the total productivity growth in the US from 1987 to 2007 can be attributed to productivity spillovers from inward FDI.
We undertake the same exercise using data on FDI from countries classified by the World Bank as low- or middle-income. Our results show that productivity spillovers exist even when the investment is from a less developed country. For this type of FDI, the impact of foreign employment shares lagged one period are negatively associated with the productivity growth of domestic firms. However, the effect is positive after a two year lag. This suggests that the spillover effects of FDI from low- to middle-income countries take longer to materialise than those of high-income countries. However, there is every reason to expect positive spillover effects of FDI from all source countries – even those that are relatively low-income compared to the US.
Conclusions and policy implications
Our analysis shows that foreign multinational firms that invest in the US are – alongside US-headquartered American multinationals – the most productive and highest-paying segment of the US economy. They conduct more R&D and provide more value-added to US domestic inputs than other firms in the American economy. The superior technology and management techniques they employ spill over in both horizontal and vertical directions to improve the performance of local firms and workers.
In an era in which the US wants not just to expand employment but to create well-paying jobs that will reverse the falling earnings that many American workers and middle-class families have suffered over recent decades, it is more important than ever to take those measures – with regard to modernising infrastructure, improving education, expanding high-skill immigration, and reforming corporate taxation – that are needed to make the US more attractive as a destination for multinational investors.
Keller, Wolfgang and Stephen R Yeaple (2009), “Multinational Enterprises, International Trade, and Productivity Growth: Firm-Level Evidence from the US”, Review of Economics and Statistics, 91(4): 821–831.
Moran, Theodore H and Lindsay Oldenski (2013), “Foreign Direct Investment in the US: Benefits, Suspicions, and Risks with Special Attention to FDI from China”, Peterson Institute for International Economics Policy Analyses in International Economics 100, August.
OECD (2013), FDI in Statistics 2013, Paris.
Olley, S and A Pakes (1996), “The Dynamics Of Productivity in the Telecommunications Equipment Industry”, Econometrica, 64: 1263–1297.