GLOBALIZATION AND PROFITABILITY. THE CASE OF ECUATORIAN FIRMS Francisco J. Guerra-Procel (Universidad Central del Ecuador) Joan Martín-Montaner (Universitat Jaume I de Castelló and Instituto de Economía Internacional) Mari D. Parra-Robles (Generalitat Valenciana. Direcció General d Internacionalització) VERY EARLY DRAFT. PLEASE DO NOT QUOTE I. Introduction There is a huge literature that has analysed, both theoretically and empirically, the role of productivity and firms being engaged into international trade. Some of these papers have found the existence of productivity premium for exporting firms. In the last decade, several papers have focused in the relationship between the globalization of firms and the impact on their profitability. Thus, Grazzi (2009) and Fryges and Wagner (2010) analyse the existence of premium on profitability for the case of firms engaged into international trade with regard to those focused on the local market. In the present study, we analyse the existence of a premium on firms benefits depending on the way they access to international trade. Our database is the Servicio de Rentas Internas (Internal Revenue Service, IRS) from Ecuador encompasses information of the whole population of Ecuadorian firms that have exported, imported or both during the period 2000-2011. We study the effect of the way these firms participate in international trade flows on their return on assets (ROA) and equity (ROE). We also take study the decomposition of ROA between the profit margin (PM), the assets turnover (AT) and the leverage (LE). We exploit the panel nature of our database. Our first results point out that returns on assets are higher for those firms that perform both as exporter as importer than for those that just focus on any of the two activities separately. But this effect mostly refers to better profit margins, whereas the assets turnover is not sensitive to the globalization of firms. Returns on equities only improve for globalized firms in specific sectors.
II. Background. Since Bernard and Jensen (1995), a huge amount of research has related firm s productivity and the exporting performance of firms. According to these authors, exporting firms are bigger in size, more capital intensive and more technologically advanced. To take into account this link, two hypotheses are usually considered: (i) selfselection of firms and (ii) learning-by exporting. The self-selection hypothesis implies a cost of entry (modification of the products, distribution ) that less productive firms cannot overcome in international markets. This is the main reason because of the most productive firms become exporters (Melitz, 2003). However, the learning-by-exporting approach means that the fact of becoming exporter implies some positive effects for firms, such as the knowledge of competitors, the tastes of consumers, new designs and strategies that translate into higher productivity. Importing firms, on the other hand, face similar entry costs to those corresponding to the exporting firms. Thus, a firm that has started any of the two activities, the entry costs for the other one are likely to decrease and makes easier to engage into it (Kasahara and Lapham, 2013). Besides, firms using imported inputs benefit from increases in productivity (Markusen, 1989). Melitz and Ottaviano (2008) show that trade liberalization favours productivity, but also firm s margins. The literature relating international trade and productivity is, therefore, large and exhaustive. Another line of research that is less extended and it is still progressing relates international trade with the role of financial markets and firm s profitability. In fact, productivity and profitability have been linked in the international trade literature: Fryges and Wagner (2010), Foster, Haltiwanger and Syverson (2008) or Bernam and Héricourt (2010). Two types of analysis can be found. On the one hand, some papers focus on the impact of the financial system on the likelihood of engaging into international trade. In this line of research, non surprisingly, it is observed that most of firms in international trade come from countries with developed financial markets and easier access to credit: Beck (2002), Goksel (2012) or Pérez and McNeil (2013) On the other hand, another group of papers focus on the impact of international trade on the financial outcome of firms. Thus, Lu and Beamish (2001) find that export intensity affects negatively profitability (whereas foreign direct investment increases it) of small and medium-size Japanese firms. Greenaway et al (2007) observe that British exporting firms enjoy healthier financial conditions than non-exporting firms. Grazzi (2012), however, do not find a premium of exporting activities on firm s profitability.
III. Database The Ecuadorian Internal Revenue Service (SRI, Servicio de Rentas Internas in Spanish) provides our database. This service records all international trade transactions in the country for fiscal purposes. Our database encompasses all firms in Ecuador involved in international transactions, as exporters, importers or both since 2000 to 2011. Thus, we have information corresponding to the whole population of firms engaged in international trade in Ecuador, not just a sample. However, this information is confined to financial issues (basically data provided by firm s income statements and corporate assets). We also know the geographical location of the firm and its type of activity, but we do not know economic information such as the number of employees. Besides, we do not have information about those firms that are not exporting or importing. Ecuador becomes an officially dollarized country in 2000, which implies this constitutes a very specific year (in 2000 the inflation rate in Ecuador was 91%, reducing to 9,3% in 2002. Thus, we will use data from 2002 to 2011. In Table 1, we present the profile of the firms in the database across time, depending on the type of trade flows recorded: firms may be exporters, importers, or they may be engaged into the two types of flows simultaneously.. Table 1 Trade profile of Ecuadorian firms across years Year Exporter Importer Both Total Number of firms % Number of firms % Number of firms % Number of firms 2002 126 7,37 690 40,37 893 52,25 1709 100 2003 139 7,6 725 39,62 966 52,79 1830 100 2004 150 7.78 753 39,04 1026 53,19 1929 100 2005 158 7,76 779 38,28 1098 53,96 2035 100 2006 178 8,16 808 37,03 1196 54,81 2182 100 2007 185 7,98 841 36,27 1293 55,76 2319 100 2008 193 8,02 864 35,91 1349 56,07 2406 100 2009 203 8,02 886 34,99 1443 56,99 2532 100 2010 204 7,88 888 34,29 1498 57,84 2590 100 2011 206 7,87 889 33,94 1524 58,19 2619 100 Total 1742 7,89 8123 33,67 12286 55,46 22151 100 Source: SRI (2012). %
The geographical distribution of firms is depicted in Table 2. As we can observe, two provinces concentrate most of the firms: Pichincha and Guayas. Table 2 Trade profile of Ecuadorian firms across provinces Province Total Importers Exporters Both Importers Exporters Both Number of firms Share Azuay 1243 711 106 423 0.57 0.09 0.34 Bolívar 13 13 0 0 1 0 0 Cañar 60 21 9 30 0.35 0.15 0.5 Carchi 39 34 0 5 0.87 0 0.13 Cotopaxi 126 81 9 36 0.64 0.07 0.29 Chimborazo 92 48 24 20 0.52 0.26 0.22 El oro 245 149 27 69 0.61 0.11 0.28 Esmeraldas 189 121 6 62 0.64 0.03 0.33 Guayas 7629 4202 523 2904 0.55 0.07 0.38 Imbabura 167 94 9 64 0.56 0.05 0.38 Loja 139 92 10 37 0.66 0.07 0.27 Los Ríos 110 31 21 58 0.28 0.19 0.53 Manabí 819 259 208 352 0.32 0.25 0.43 Pastaza 20 0 10 10 0 0.5 0.5 Pichincha 10284 5931 691 3662 0.58 0.07 0.36 Tungurahua 625 323 60 242 0.52 0.1 0.39 Galápagos 6 6 0 0 1 0 0 Orellana 11 11 0 0 1 0 0 Santo domingo de los Tsáchilas 170 84 22 64 0.49 0.13 0.38 Santa Elena 100 37 3 60 0.37 0.03 0.6 Source: Servicio de Rentas Internas (2012). IV. Financial profile of firms In our database the information regarding firm s accounting balance is available only at an aggregate level. However, we are able to construct the standard RAO (Returns on Assets) and ROE (Return on Equity) indicators. Return on Assets The Return on Assets index (ROA) allows knowing about firm s efficiency in the management its resources and the profitability of the firm s investments, regardless of the way they have been financed: ROA = Net Income Total Assets Returns on Assets can be caused by two effects: the profit margin that a firm can obtain from its sales (PM) and the assets turnover (AT), which measures the efficiency in the firm s use of assets to generate sales income. Therefore, we can write the ROA index as
Net Income Net Income Sales ROA = = Total Assets Sales Total Assets The first component stands for the profit margin (PM) whereas the second one proxies the assets turnover (AT). Return on Equity The Return on Equity (ROE) measures how well the firm generates income from the equity available to it: ROE = Net Income Shareholder Equity In this case, we can also provide additional information by incorporating the financial leverage of the firm. Thus, the following decomposition of the ROE index ROE = Net Income Net Income = Shareholder Equity Sales Sales Total Assets Total Assets Shareholder equity includes the NP and the AT terms previously defined plus the third element, which measures the share of the assets being financed by equity and, therefore, measures indirectly the level of indebtedness of the firm (FL): the higher the value of this term, the higher the weight of debts in the balance sheet of the firm. Table 3 Financial Ratios by type of firm (averages) ROA ROE NP AT FL Exporter 0,0370 0,157 2,786 0,034 12,176 Importer 0,0469 0,268 14,33 0,027 23,593 Both 0,0565 0,218 1,655 0,037 15,819 Source: Servicio de Rentas Internas (2012). V. Specification We estimate the following model: Where: lnf!"# = α! + α! d!"!"# + α! d!"#!"# + γ! + φ! + δ! + ε!,!
ln F!"# is a vector of financial characteristic of a firm in sector i, located in the province k in year t. The financial characteristics are measured through the previous indicators: returns on assets (ROA) and returns on equity (ROE). Our key explanatory variables are these two dummies:!" = Binary variable that takes value 1 for firms that only export (x), and 0 otherwise (that is, also import or only imports).!"# = Binary variable that takes value 1 for firms that simultaneously exporter and importer (mx), and 0 otherwise (she only exports or imports). Therefore, those firms that only imports integrate our group of control. We also include some fixed effects to take into account (i) non-observed effects that are specific to the type of activity and the location of the firm that do not change across time and (ii) effects that are common to all activities and locations but varying across time. The two former could indicate the existence of comparative and locational advantages (respectively), whereas the latter controls for the economic cycle. All of them also stand as control for likely errors of measurement. The set of indicators is: γ! = Industry fixed effects. φ! = Province fixed effects δ! = Year fixed effects. If we take logarithms, decomposition of the indicator of return on assets would be expressed as: ln ROA = ln PM + ln AT This exercise allows us to analyse the effect of international trade on both components of the returns. Obviously, the same exercise can be done in the ROE indicator: ln ROE = ln PM + ln AT + ln FL VI. Results In Table 4 we report the estimates for the ROA and ROE indicators as dependent variables. With regard to the return on assets, we observe the existence of a Premium for those firms which simultaneously are involved in exporting and importing activities with
regard to those which are just importing. This first result fits well with the theoretical literature that as the literature recognizes exporting firms as more efficient than not exporting firms. However, this profitability premium turns out to be negative for those firms that are only exporters. This latter result, however, is only achieved when we do not include in our estimates sector fixed effects (columns (1) and (2)). This outcome suggests that this negative premium to exporter firms presents some sector specifics that should be explored. In the case of the return on equity, no premium is observed for exporter-and-importer firms, whereas the negative premium for exporter firms follows a similar pattern to the obtained for the returns on assets.
Table 4. Existence of trade Premium on Firm Returns Random effects panel ln ROA ln ROE VARIABLES (1) (2) (3) (4) (5) (6) (7) (8)!"!"# - 0.007* - 0.007* - 0.00348-0.00351-0.0301** - 0.0287* - 0.0238-0.0232 (0.004) (0.00371) (0.00370) (0.00375) (0.0150) (0.0151) (0.0152) (0.0154) 0.013*** 0.013*** 0.0140*** 0.0141*** 0.00880 0.00972 0.0106 0.0115 (0.002) (0.002) (0.00210) (0.00211) (0.00849) (0.00851) (0.00862) (0.00864) Constant 0.011*** 0.004 0.00253-0.00498 0.0551*** 0.0392** 0.242*** 0.220** (0.002) (0.004) (0.0223) (0.0227) (0.00725) (0.0176) (0.0918) (0.0933) Specific Effects Year Yes Yes Yes Yes Yes Yes Yes Yes Sector No No Yes Yes No No Yes Yes Province No Yes No Yes No Yes No Yes Observations 21,757 21,757 21,757 21,757 21,036 21,036 21,036 21,036 Number of id 2,741 2,741 2,741 2,741 2,704 2,704 2,704 2,704 rho 0.377 0.378 0.373 0.374 0.441 0.440 0.440 0.439 p 0 0 0 0 0 0 0 0 chi2 3377 3393 3431 3445 2884 2909 2910 2933 r2_w 0.152 0.152 0.152 0.152 0.124 0.124 0.124 0.124 r2_b 0.00449 0.00735 0.0162 0.0197 0.0942 0.103 0.102 0.110 r2_o 0.0950 0.0972 0.102 0.105 0.112 0.115 0.115 0.118 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
Next, we analyse in Table 5 whether we can find a specific premium for those firms located in the two main provinces in the country (Pichincha and Guayas) by introducing two additional dummy variables: prov_exp and prov_both, depending on the firms located in those regions being just exporter or both importer and exporter. We also include specific dummies for the firms in the two sectors with the highest number of firms engaged in international trade: (i) Manufacture of food products and beverages and (ii) Manufacture of Chemical Substances and Products: inds_exp and inds_both. We do not find any premium on revenue, on assets or on equity, depending on the geographical location of the firms, as seen in columns (1) and (3) in Table 2. However, the type of activity reveals that exporting firms operating in food products and chemicals show, in fact, higher returns than those that are only importing. Thus, in the case of exporting firms, the net effect on ROA is positive although small: 0.009 + 0.016 = 0.007, whereas the premium for firms importing and exporting is higher than the one obtained for firms in other sectors: 0.01 + 0.011 = 0.21. The same pattern is observed for the ROE indicator (see columns (2) and (4)). Table 5. Existence of trade Premium on Firm Returns. Sectorial and regional differences Random effects panel ln(roa) ln(roe) VARIABLES (1) (2) (3) (4)!" - 0.011-0.009** - 0.013-0.046**!"# (0.007) (0.005) (0.030) (0.019) 0.007 0.010*** - 0.009-0.002 (0.005) (0.002) (0.020) (0.011) prov_exp 0.009 - - 0.016 - (0.008) (0.035) prov_both 0.008-0.026 - (0.005) (0.022) inds_exp - - - 0.016** - - - 0.062** (0.008) (0.031) inds_both - - - 0.011** - - - 0.039** (0.005) (0.018) Constant 0.006-0.001 0.047** 0.027 (0.005) (0.005) (0.021) (0.021) Specific Effects Year Yes Yes Yes Yes Sector Yes Yes Yes Yes Province Yes Yes Yes Yes Observations 21,757 21,757 21,036 21,036 Number of id 2,741 2,741 2,704 2,704 Rho 0.374 0.373 0.439 0.439 P 0 0 0 0 chi2 3448 3454 2935 2941 r2_w 0.152 0.152 0.124 0.124 r2_b 0.0204 0.0223 0.111 0.113 r2_o 0.105 0.106 0.118 0.118 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
Next, we analyse the effect of being an exporter firms on the different components of ROA and ROE that we can identify through the decomposition explained above: ln ROA = ln PM + ln AT ln ROE = ln PM + ln AT + ln FL The results are displayed in Table 6. One interesting feature is that the premium observed on the ROA for exporting/importing firms comes from the proft margin only (column 1), with no effect on the assets turnover (column 2). When we extend the analysis to the ROE and include the financial leverage (column 3), a negative premium appears for all firms engaged in exporting activities. In other words, exporting firms are less indebted than importing firms. Table 6. Existence of trade Premium on Firm Returns. Return components Random effects panel ln(roe) ln(roa) ln(pm) ln(at) ln(fl) VARIABLES (1) (2) (3)!" -0.00191 0.0171-0.158*** (0.00267) (0.0305) (0.0606)!"# 0.0107*** -0.00930-0.136*** (0.00147) (0.0173) (0.0342) Constant -0.00250 0.797*** 2.834*** (0.0156) (0.185) (0.367) Specific Effects Year Yes Yes Yes Sector Yes Yes Yes Province Yes Yes Yes Observations 20,182 21,064 20,817 Number of id 2,667 2,740 2,700 Rho 0.397 0.531 0.539 P 0 0 0 chi2 2715 197.2 166.8 r2_w 0.126 0.00239 0.00469 r2_b 0.0610 0.0523 0.0330 r2_o 0.103 0.0345 0.0186 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 In Table 7 our previous analysis taking into account the main provinces in Ecuador and the most globalised sectors is reproduced for the decomposition of returns. We observe that being and exporter/importer firm implies a significantly higher profit margin only in Pichincha and Guayas (column 1). In those provinces, the level of indebtedness is smaller only for exporter firms, whereas exporter/importer firms display lower levels of indebtedness all over the country (column 3). In our analysis by sectors, exporter firms present a smaller margin profit than importing firms except in the food and chemicals sectors (column 4). Also in these industries, the assets turnover is better for all exporting firms.
Table 7. Existence of trade Premium on Firm Returns. Return components with sectorial and regional differences Random effects panel ln(roe) ln(roe) ln(roa) ln(roa) ln(pm) ln(ra) ln(fl) ln(pm) ln(ra) ln(fl) VARIABLES (1) (2) (3) (4) (5) (6)!" 0.001-0.005 0.038-0.006* -0.025-0.124 (0.005) (0.059) (0.119) (0.003) (0.039) (0.077)!"# 0.005 0.041-0.147* 0.010*** -0.035-0.114*** (0.003) (0.041) (0.080) (0.002) (0.022) (0.043) prov_exp -0.004 0.035-0.270** --- --- --- (0.006) (0.069) (0.138) prov_both 0.007* -0.062 0.015 --- --- --- (0.004) (0.045) (0.089) inds_exp --- --- --- 0.012** 0.117* -0.095 (0.006) (0.062) (0.124) inds_both --- --- --- 0.001 0.071** -0.061 (0.003) (0.036) (0.0714) Constant 0.012*** 0.811*** 1.760*** 0.009*** 0.801*** 1.788*** (0.00361) (0.0416) (0.0829) (0.004) (0.041) (0.081) Observations 20,182 21,064 20,817 20,182 21,064 20,817 Number of id 2,667 2,740 2,700 2,667 2,740 2,700 rho 0.397 0.531 0.538 0.397 0.530 0.539 p 0 0 0 0 0 0 chi2 2720 200.0 171.3 2720 203.6 167.8 r2_w 0.126 0.00239 0.00469 0.126 0.00239 0.00469 r2_b 0.0622 0.0534 0.0350 0.0625 0.0541 0.0333 r2_o 0.104 0.0352 0.0190 0.103 0.0362 0.0188 Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
Concluding Remarks References Beck, T. (2002). Financial development and international trade: Is there a link? Journal of International Economics, 57 (1), 107 131. Bernam N., and Héricourt, J. (2010). Financial factors and the margins of trade: Evidence from cross-country firm-level data, Journal of Development Economics, 93, 206 217. Bernard, A.B. and Jensen, J. (1995). Exporters, Jobs, and Wages in U.S Manufacturing: 1976-1987, Brookings Papers on Economic Activity: Microeconomics, 67-119. Foster, L., Haltiwanger, J. and Syverson, C. (2008). Reallocation, Firm Turnover, and Efficiency: Selection on Productivity or Profitability? American Economic Review, 98 (1), 394-425. Fryges, H. and Wagner, J. (2010). Exports and profitability: First evidence for german manufacturing firms, The World Economy, 33(3), 399.423 Goksel, T. (2012), Financial Constraints and International Trade Patterns. Economic Modelling, 29, 2222-2225 Grazzi, M. (2012). Export and Firm Performance: Evidence on Productivity and Profitability of Italian Companies, Journal of Industrial Competition and Trade, 12, 413-444. Greenaway D, Guarigla, A. and Kneller, R. (2007). Financial Factors and Exporting Decisions. Journal of International Economics, 73 (2), 377 395. Kasahara, H., and Lapham, B. (2013). Productivity and the decision to import and export: Theory and evidence, Journal of International Economics, 89 (2), 297-316. Lu J.W. and Beamisch, P.W. (2001): The Internationalization and Performance of SMEs, Strategic Management Journal, 22 (6-7), 565-586. Markusen, J. R. (1989), Trade in Producer Services and in Other Specialized Intermediate Inputs, American Economic Review, 79 (1), 85 95. Melitz, M.J. (2003). The Impact of Trade on Intra-Industry Reallocations and Aggregate
Industry Productivity. Econometrica, 71(6): 1695 1725. Melitz, M. J. and Ottaviano, G. I. P. (2008). Market size, trade, and productivity, Review of Economic Studies 75(1), 295 316. Pérez D. and McNeil, L. (2013). The impact of trade finance on international trade: Does financial development matter? Research in Business and Economics Journal. 8 (1), 1-19