VoxEU Column International trade

Do ‘animal spirits’ matter to firms’ internationalisation?

How do firms go international? This column reviews evidence from industrialised and emerging economies, including Japan and China, with some surprising findings.

The internationalisation of firms’ production activities is having a massive impact on the global economy – everything from facilitating the rapid industrialisation and income growth in China and other emerging economies to the hollowing out of G7 manufacturing sectors. This growth and de-industrialisation is, in turn, blamed for booming commodity prices and rising wage inequality. Plainly, understanding the determinants of firms’ internationalisation is critical to comprehending today’s globalisation.

The driving forces behind firm internationalisation, such as export, foreign direct investment (FDI), and offshoring, have been much studied. Theoretically, heterogeneous-firm models of trade such as those in Melitz (2003) and Helpman et al. (2004) predict positive effects of productivity and firm size on firm internationalisation. That is, only productive and large firms can become exporters and FDI firms, since internationalisation is associated with initial sunk costs. Many empirical studies using firm-level data found evidence supporting this prediction (see Clerides et al. 1998, Bernard and Jensen 1999 and 2004, and Greenaway and Kneller 2004, among many others).

Looking beyond firm-level productivity

Yet some of these empirical studies find that the effect of productivity is small in size. For example, using US plant-level data, Bernard and Jensen (2004) find that a 100% increase in total factor productivity (TFP) raises the probability of exporting by only 1.7 percentage points. Bernard and Wagner (2001) find similar-sized effects of labour productivity on export decisions using German data. My recent work, which uses firm-level data for Japan finds a negligible effect of productivity: a 50% increase in productivity raises the probability of engaging in export or FDI by less than 0.1 percentage points (Todo 2011). These findings suggest that factors other than productivity may be important in determining firms’ internationalisation.

Risk and time preference of firms may be one of such factors. In recent work with Hitoshi Sato, I take the advantage of a rich firm-level data set for Japanese small and medium enterprises (SMEs) and find that SMEs are more likely to be internationalised when the firm’s CEO is more risk-tolerant, forward-looking, and internationally experienced (Todo and Sato 2011). These effects are large in size, much larger than the effect of productivity.

Evidence from privatisation of Chinese firms

More recently, in work with colleagues, I look into this issue from a different perspective, utilising evidence from privatisation of Chinese state-owned enterprises (SOEs) (Todo et al. 2012). SOEs in China have been gradually privatised since the early 1990s, as the central government started to be concerned about large debts of SOEs. Accordingly, the number of SOEs drastically declined from about 23,000 in 2000 to 5,000 in 2007.

Using firm-level data for China, we find a large positive effect of privatisation on exporting decisions. When non-exporting SOEs are privatised, their probability of initiating export in the next year increases by 2.1 percentage points on average whereas the average probability is 2.4%. Also, privatisation improves the productivity (TFP) level and the firm size (the number of workers) in the next year by 3.3 and 2.8 percentage points, respectively, on average (See Figure 1 below). Thus, the improvement in productivity and size could be channels of the positive effect of privatisation on exporting decisions.

Figure 1. Effects of privatisation of Chinese state-owned enterprises

Note: These figures indicate the point estimate of the effect of privatisation (circles) and its 95%-level confidence intervals.

However, although productivity and firm size positively affect firms’ exporting decisions, the size of these effects was found to be small in the case of China, as is often the case in existing studies for other countries. For example, the coefficient on the log of productivity in the linear probability estimation for exporting is 0.015, similar to the result in Bernard and Jensen (2004) for the US. Since privatisation increases the productivity by 3.3%, we can conclude that privatisation raises the probability of exporting by 0.052 percentage points (=3.3% * 0.015) through the productivity improvement. Since privatisation increases the probability of exporting by 2.1 percentage points in total, the effect through productivity improvement seems to be negligible. Similarly, the effect through improvement in firm size is also negligible.

These results imply that privatised SOEs are more likely to engage in export than remaining SOEs, mostly due to unobservable factors, which cannot be explained by productivity or firm size. A potential candidate of such unobservable factors is changes in firms’ attitude toward profits and risks. When privatised, former SOEs are not protected by the government any more so that they may have to expand their business to survive. One way to expand it is to export to foreign markets. Moreover, privatised SOEs should become more risk-taking to be profitable than SOEs and may thus initiate exporting. My earlier finding with Hitoshi Sato indeed finds that more risk-taking firms are more likely to export and that the effect of risk preference on exporting decisions is far more important in magnitude than the effect of productivity and firm size, using firm-level data for Japanese SMEs. These psychological factors may play an important role in effects of privatisation.


As is almost always the case, we need more evidence before reaching a final conclusion. The new research, however, indicates that ‘animal spirits’ matter to firms’ internationalisation.

If this is in fact the case, these findings can provide an implication to policies for export promotion, which has been an important target for policy makers in many countries. In the US, for example, President Obama announced the National Export Initiative in 2010 to double US exports by 2014. Export promotion policies for firms generally include supporting export promotion efforts (e.g., participation of firms in overseas exhibitions and expansion of export promotion agencies), subsidising production, and increasing access to credit.

The effectiveness of such policies is ambiguous, according to recent empirical studies using firm-level data. Some find a positive effect of export promotion policies (Görg et al. 2008; Martincus and Carballo 2008; and Van Biesebroeck et al. 2010), while others find an insignificant effect (Bernard and Jensen 2004; Girma et al. 2009). Therefore, we need more knowledge on how policies can promote exports more effectively (assuming that such policies are welfare enhancing possibly due to spillovers of information about export across firms).

Recent research findings suggest that stimulating animal spirits – for example by providing information about foreign markets, helping to expand networks with foreign firms, and expanding the coverage of trade insurance – may be a key to effective export promotion.

Editors' Note: The main research on which this column is based (Todo et al. 2012) first appeared as a Discussion Paper of the Research Institute of Economy, Trade and Industry (RIETI) of Japan.


Bernard AB and JB Jensen (1999), “Exceptional exporter performance: Cause, effect, or both?”, Journal of International Economics, 47:1-25.

Bernard AB and JB Jensen (2004), “Why some firms export”, Review of Economics and Statistics, 86:561-569.

Bernard AB, J Wagner, (2001), “Export entry and exit by German firms”, Weltwirtschaftliches Archiv, 137:105-23.

Clerides SK, S Lach, JR Tybout (1998), “Is learning by exporting important? Micro-dynamic evidence from Colombia, Mexico, and Morocco”, Quarterly Journal of Economics, 113:903-947.

Girma, S, H Gorg, and J Wagner (2009), “Subsidies and exports in Germany: First evidence from enterprise panel data”, IZA Discussion Paper, No. 4076.

Gorg, H, M Henry, and E Strobl (2008), “Grant support and exporting activity”, Review of Economics and Statistics, 90(1).

Greenaway D and R Kneller (2004), “Exporting and productivity in the United Kingdom”, Oxford Review of Economic Policy, 20:358-371.

Helpman E, MJ Melitz, SR Yeaple (2004), “Export versus FDI with heterogeneous firms”, American Economic Review, 94:300-316.

Martincus, Christian V and Jeronimo Carballo (2008), “Is export promotion effective in developing countries? Firm-level evidence on the intensive and the extensive margins of exports”, Journal of International Economics, 76(1).

Melitz MJ (2003), “The impact of trade on intra-industry reallocations and aggregate industry productivity”, Econometrica, 71:1695-1725.

Todo Y (2011), “Quantitative evaluation of determinants of export and FDI: Firm-level evidence from Japan”, The World Economy, 34:355-381.

Todo Y, T Inui, and Y Yuan (2012), “Effects of Privatization on Exporting Decisions: Firm-level evidence from Chinese state-owned enterprises”, RIETI Discussion Paper, No. 12-E-015.

Todo Y and H Sato (2011), “Effects of CEO’s characteristics on internationalization of small and medium enterprises in Japan”, RIETI Discussion Paper, No. 11-E-026.

Van Biesebroeck, J, E Yu, and S Chen (2010), “The impact of trade promotion services on Canadian exporter performance”, available at SSRN: http://ssrn.com/abstract=1612209 or http://dx.doi.org/10.2139/ssrn.1612209

630 Reads