Spain’s problems of high unemployment and soaring sovereign debt are well known. What is less publicised is Spain’s poor export performance. The numbers tell the story; only 6% of Spain’s manufacturing exports are high-tech, compared with 15% in Germany and 25% in France1. While Spain cannot hope to solve such a deep-seated problem as its trade orientation overnight, it may be useful to demonstrate, by way of a thought-experiment, the benefits to a group of exporting Spanish firms (or ‘export starters’) compared to a control group of never-exporting Spanish firms.
How do export-oriented firms improve?
What we want to demonstrate is that there are additional benefits to exporting that go beyond obvious positive aspects such as selling additional output and earning export income. Iacovone and Javorcik (2010) have recently shown how Mexican firms that are due to export upgraded their products in order to be able to compete on international markets. What has not been studied so far is whether firms improve the manner in which they manufacture goods, that is, the export orientation’s effects on ‘process changes’. The expectation is that a newly exporting firm introduces improvements either in anticipation of foreign competition (‘product upgrading’) or in response to the export experience (‘learning’).
It seems intuitive that upgrading and learning takes place when firms aim to export. Yet, hard evidence is difficult to come by because of this topic’s unsuitability to econometric regression. Firms that export tend to already be better than non-exporters at most other activities, leading to the econometric problem of incorrect attribution. As Joachim Wagner (2002) put it, ‘[i]f today’s export starters are ‘better’ than today’s non-exporters (and have been so in the recent past), we would expect that they should, on average, perform better in the future even if they do not start to export today’.
How to tackle the ‘exporters-are-generally-better’ effect
How can we rule out the ‘exporters-are-generally-better’ effect? We have dealt with this problem by focusing on improvements for a subset of ‘export starters’. ‘Export starters’ are firms that have switched to exporting having not exported before. We compare improvements for this subset of firms against changes registered for a control group. The control comprises firms that have not exported since 2006 and have remained export-inactive. To help us compare, we use a technique known as Propensity Score Matching (PSM)2, a procedure similar to the Nearest Neighbour technique. Propensity Score Matching involves pairing ‘never exporting firms’ with firms that have switched to exporting. We can do this by empirically identifying exporting and non-exporting firms with their roughly-equal chances of becoming exporters. Only 6% of firms in our sample transition from a non-exporting to an exporting mode. This small percentage matches similar research in this field (e.g. Girma et al. 2003).
The most efficient and smartest firms are those most likely to cover the sunk costs of transitioning to export mode. Our technique thus concentrates on those that switch to exports We can.then match these firms with ‘never exporting’ firms who are predicted to be equally as likely to become exporters based on their comparable efficiency, size, age and the level of technology used. Now we have narrowed down the only difference that separates the two groups; any changes in innovation performance between 2004 and 2005 when firms leave the ‘never exporting’ group and become exporters. We can then further refine our study by using Difference-in-Differences: comparing not only the differences between the two groups, but also changes in these differences over time. This helps us control for external events that help and hinder all firms equally.
Exporters improve their process innovations
Export starters improve the way they do things. Not the product. The only significant difference in self-reported innovation rates is for process innovations at the time of market entry. Our results show that ‘export starters’ report 11% higher innovation rates (Figure 1). .
Figure 1 Derived self-reported innovation rate margins for ‘export-starters’ (vs. never- exporting)
We deem product innovation to be of little importance. First, only 7% of all firms claim to have made any improvements in product innovation in 2005. Second, process innovation rates are at least twice this value. Therefore any changes to the number of self-reported process innovations are subject to considerably less random variation than changes to levels of self-reported product innovation.
When this paper was recently presented at a workshop hosted by Statistics Sweden, a discussant claimed that our finding regarding post-entry process innovations was a “non-result”. We politley disagree. ‘Export starters’ improved their production processes in the entry year of 2006 and these improvements had significant knock-on effects. Our findings are consistent with similar work by Girma (2003), who found that for exporters ‘the benefits from exporting are due to exposure to best practice technology, rather than scale economies or competition effects”.
Are these improvements merely cosmetic? We cannot exactly discern from our data, but the effects of technological persistence that have been proposed by Girma are likely. Our analysis - which uses data from the Spanish Business Strategy Survey (SBSS) for 2004 to 2006 - suggests significant real world benefits for Spanish firms orienting themselves towards exporting. Such a change could well help these firms to employ a more imaginative and cost-efficient approach to producing and selling their products in a time of economic woe.
Ferro, E (2011), “Signaling and Technological Marketing Tools for Exporters”, World Bank Policy Research Working Paper, 5547.
Girma, S, D Greenaway and R Kneller (2003), “Export Market Exit and Performance Dynamics: A Causality Analysis of Matched Firms”, Economics Letters, 80(2), 181–187.
Hanley, A and J Monreal-Pérez (2012), “Are Newly Exporting Firms more Innovative? Findings from Matched Spanish Innovators”, Economics Letters, 116(2), 217–220.
Iacovone, L and B Javorcik (2010), “Multi-Product Exporters: Product Churning, Uncertainty and Export Discoveries”, The Economic Journal, 120(544), 481–499.
Wagner, J. (2002), “The Causal Effects of Exports on Firm Size and Labor Productivity: First Evidence from a Matching Approach”, Economic Letters, 77, 287-292.
1 From World Bank Development Indicators, 2010.
2 Cf. Hanley and Monreal-Pérez (2012) for a more extended discussion of the methodology.