Why the best exporters are not necessarily the best producers

The beneficial effects of trade on the economy depend mostly on the fact that trade enhances competition and favours resource reallocation towards the best firms. Promoting exports is then a policy that allows countries to improve their economic performance, boost GDP growth, and create new and better jobs. This policy, however, can be effective only to the extent that it correctly identifies the high-flying firms as well as their skills.

Bernard et al. (2012) discovered an interesting characteristic of firms’ export behaviour, i.e. the fact that a large share of exports consists of goods not produced by the firm which sells them, and named it ‘carry along trade’ (CAT).1

• My new research on this characteristic reveals that very often the best exporters are not necessarily the best producers (Di Nino 2015).

A large share of their exports are sourced products, and their main competences are marketing and the ability to access selling networks, rather than crafting higher quality or cheaper products.

Let me make a few well-known examples. A coffee machine sold by Nespresso at its stores, but entirely produced by De’Longhi or Krups, is a carry along trade good. Similarly, the motorbike maker Triumph involves in carry along trade when it sells apparels – which it has not produced – in its stores under its brand. Spare parts of cars can also be carry along trade goods if they are bought and resold as final goods, without any additional transformation, not necessarily under the carmaker brand. By contrast, the same spare parts are intermediate inputs if they are used as components.

The results of specific questions included in the Survey on the Italian Industrial and Services Firms in 2012 reveal that carry along trade concerns around a quarter of the Italian domestic manufacturing, as well as of export sales, and is performed by 36% of the exporters. The majority of carry along exports are sold abroad under the final seller trademark (almost 60%). Their fabrication origin is then concealed to final customers, who (may) perceive sourced goods as the own production of the firms that sell them.

Features of carry along trade firms

Firms involved in carry along trade have the same features normally associated with the ‘happy few’ (Mayer and Ottaviano 2007). In particular, they are big companies, pay higher salaries, invest more abroad and in R&D, and have fewer financial constraints. This is because they are generally exporters (9 out of 10) and about half of them are multinationals. The share of exports on total sales is larger for carry along trade firms than for other exporters (Figure 1a), but these larger foreign sales are entirely due to sourced goods; the share of own production that is exported is indeed the same for carry along and non-carry along exporters (around 35%).

So the question naturally arises is to what advantage carry along firms get from trading other firms’ products. According to the survey, firms engage in such activities mainly to lower production costs or, less frequently, to exploit synergies arising from the joint sale of sourced and internally produced goods. Moreover, what such firms mostly do in order to sell sourced goods is actually logistic and distribution (almost 70%), packaging (19%), and only for a small quota R&D/product design. Thus, it is a specific advantage in servicing costumers, especially on foreign markets, that seems to give carry along firms their competitive edge, and makes it convenient or necessary for other firms to sell their products through them.

Selling patterns and task specialisation of firms engaged in carry along trade evoke the concept of ‘buyer-driven global value chains’, i.e. production chains where the leading firm specialises in downstream activities with high value added. At the same time, carry along firms are an abridged version of a manufacturing and a service enterprise, still producing goods – thus retaining a manufacturing ‘status’ – but providing complementary services for other manufacturers – although not fully specialising in those – constituting true archetypes of the manufacturing-servicification process.

Open issues on firm productivity

Carry along trade is a widespread case of incomplete ‘task-specialisation’ as firms perform only specific tasks for some products and the entire production process for others. This phenomenon then introduces two potential biases affecting the usual measures of firm-level productivity, which may turn out to be very severe given the large number of firms involved and the high share of manufacture that is concerned.

• The activity performed by firms which produce the whole good but do not market it is very different from the activity of carry along trade firms, which instead do not produce a significant share of their sales.

Thus, the hypothesis maintained in total factor productivity (TFP) estimates that the production process is identical for every firm within the same industry is not valid. Fuss and Warzynski (2012) identify a similar issue for the TFP estimates of multi-product firms. In their case, the bias originates from the impossibility of assessing the distribution of factors across production lines. Analogously, the existence of carry along trade distorts TFP estimates because production factors cannot be assigned to separate tasks. As a result, the TFP of firms specialised in high value added stages of production appears greater, but this may reflect the different bundle of activities performed rather than a superior skill in organising the production factors within tasks.

• A specialisation in downstream activities like marketing and distribution may reveal a special ability in retaining customers.

At the same time, it enhances the visibility of the carry along brand to consumers, which in the end skews their preferences towards the products of these firms. A fact that, in turn, affects positively the price they can charge and consequently their ‘measured productivity’. Productivity estimates that fail to take into account this aspect may then confuse the proximity to final buyer with the efficiency in production.

Carry along trade therefore calls for a refinement of the TFP concept, which encompasses activities complementary to physical production of a good, while distinguishing their specific contribution.

Are the potential biases in measuring firm productivity a concrete issue?

Exploiting the fact that the productivity of companies  exporting not-own produced goods varies with the self-assessed incentive to provide ‘carry along services’ and the task accomplished by them, one can study how the productivity premium is related to carry along trade.2 Estimates show the existence of a ‘small’ productivity premium for carry along trade firms (between 2 and 3%, see Figure 1b) which gets much larger when:

• The brand of the firm is renowned (20% premium);
• Firms operate primarily in packaging (10%) and distribution (3%) on sourced goods;3
• Carry along trade concerns exports; in this case, the premium is 5% on average, and increases with the share of sourced goods on total exports.

The productivity premium associated with carry along trade activity entangles efficiency in production, selling ability (access to markets through well-established distributional channels), and possibly a certain degree of market power given by the proximity to final consumers and the marketing skills of the firms. The risk of confounding productivity with other sources of a firm’s profitability seems to be very pronounced for these firms.

Concluding remarks

In tailoring industrial and trade policies, policymakers cannot ignore that productivity does not represent simply a firm ability in combining input factors within the production process and that frequently what prevents firms from exporting is not the lack of manufacturing skills but various obstacles faced to sell their final production. If what many firms miss is the ability to sell, rather than the ability to produce, then liberalisation of trade in services can do a lot of good to manufacturing trade. The tighter and tighter interaction between these two sectors thus deserves particular attention from the policymakers involved in the negotiations of trade agreements.

Figure 1a. Domestic and foreign sales for the average non-CAT and CAT firm (share of total sales)

Figure 1b. Productivity premium of Multinationals (MNE) and CAT firms: how it varies with CAT features and activity

References

Bernard, A, E Blanchard, I Van Beveren and H Vandenbussche (2012) “Carry-along Trade”, NBER Working Paper No. 18246, July.

Borin, A and M Mancini (2015) “Foreign Direct Investment and Firm Performance: An empirical analysis on Italian firms”, Temi di Discussione, Bank of Italy Working Papers, forthcoming.

Costinot, A, L Oldenski and J Rauch (2011), “Adaptation and the Boundary of Multinational Firms”,The Review of Economics and Statistics, 93(1), 298-308.

De Angelis, E, S De Nardis and C Pappalardo (2011), “Produttori ed esportatori multiprodotto “, inItaly in the World Economy, ICE-ISTAT Report 2010-2011, ed. by ISTAT, 307 – 316.

Di Nino, V (2015), “The phenomenal CAT: firms clawing the goods of others”, Questioni di Economia e Finanza, Bank of Italy Occasional Papers , forthcoming.

Fuss, C and F Warzinsky (2012), “The Determinants of Markups on Export Markets: A firm-product-market Analysis”, mimeo.

Mayer, T and G I P Ottaviano (2007), “The happy few: the internationalisation of European firms New Facts based on firm-level evidence”, Sciences Po publications.

Footnotes

1 Carry along trade (CAT) describes the sales, at home and/or abroad, of final goods by a manufacturing firm (that also produces goods in-house), which are entirely fabricated by another manufacturing firm. Each case of CAT is indeed at the same time a situation of mediated production for the final seller and of mediated sales for the producer. Here I focus on the incentives and features of the selling firms, therefore the final seller is the CAT firm and CAT goods are products sourced from third companies. CAT does not cover services provided by distribution companies, nor does it concern intermediate inputs.

2 Total factor productivity measures are related to the status of a carry along trade firm, to the reasons for operating such firms, to the activity involving in carry along trade and to the its share in sales on total and export turnover. TFP is a firm-specific, price deflated measure of productivity, obtained from Borin and Mancini (2015), by applying standard methods à la Levinson-Petrin. Price indices, employed as deflator, are almost firm specific because sectors are very narrowly defined. The estimated equations control for time, sector, firm’s size, exporter and multinational status, therefore the carry along trade premium must be read as a specific extra productivity over the average firm which is not captured by any of the above listed factors.

3 Similar conclusions are derived from the analysis (for a small subset of companies for which information was available) of the gross profit margins, separately computed within each firm for in-house and sourced goods. Profit margins tend to be higher on carry along trade goods, especially when they involves mainly packaging, rebranding and/or distribution.

This article is published in collaboration with Vox EU. Publication does not imply endorsement of views by the World Economic Forum.

To keep up with the Agenda subscribe to our weekly newsletter.

Author: Virginia Di Nino is an Economist in the Emerging Markets and International Trade Division, Banca d’Italia.

Image: A container ship departs Burrard Inlet in Vancouver, British Columbia. REUTERS/Andy Clark