1、 外文翻译 原文 The Determinants of Growth for Small and Medium Sized Firms. The Role of the Availability of External Finance Material Source:EBSCO Author:Leonardo Becchetti,Giovanni Trovato We present an empirical analysis of the determinants of growth for a sample of Italian small and medium sized firms.
2、We show that,when investigating a sample which includes firms between 10 and 50 employees and a set of variables larger than those usually considered in the literature,growthnet of industry characteristics and ex ante market powerturns out to be significantly affected not only by size and age,but al
3、so by state subsidies,export capacity and credit rationing.By adopting a multivariate approach we also show that these findings are confirmed after controlling for heteroskedasticity,survivorship bias and serial correlation.Our results suggest that the hypothesis of independence of firm growth from
4、the initial size and other factors(usually referred to as Gibrats law in the literature)is not rejected for large firms,while it does not hold for small and medium sized firms under financial constraints in a“bank-oriented”financial system in which access to external finance is difficult. Since 1931
5、,when it was first formulated,Gibrats law has been a useful theoretical benchmark for theoretical and empirical research on the determinants of firm growth.Its two main points may be summarised as follows:(i)the rate of growth of a firm is independent from its size at the beginning of the period;(ii
6、)the probability of a given rate of growth during a specific time interval is the same for any firm within the same industry.It is worth noting that the second point is more general than the first one and it implies that,after controlling industry characteristics,the expected growth rates should not
7、 be affected by any other variable. Most empirical analyses reject the hypothesis of independence of growth from size and age.Firm growth is significantly and negatively related to size and age when only surviving firms are considered and,to a lesser extent,when survivorship bias is taken into accou
8、nt.The paucity of available data prevented investigation in other directions of interest.More detailed empirical analysis of the determinants of growth going beyond the traditional size-age-growth relationship should be of great relevance both for economists and policymakers.To consider an example,t
9、he relationship among firm size,availability of external finance,access to foreign markets and ownership structure has been largely neglected even though it may provide significant policy insights on the optimal corporate governance and regulation of financial institutions in support of industrial a
10、nd economic growth.This paper aims to fill this gap by considering a large set of potential determinants of firm growth.The paper is divided into five sections(including introduction and conclusions).In the second section we briefly describe theoretical and empirical findings showing that a signific
11、ant impact of size and age on firm growth has been found for different countries and estimation periods.In the third section we present some descriptive findings regarding the determinants of growth for a sample of about 4000 Italian firms between 1989 and 1997.In the fourth section we present our e
12、conometric results which control for heteroskedasticity,serial correlation and survivorship bias.These results show that size and age are not the only determinants of growth and that financial constraints and access to foreign markets have a significant impact on the growth of small and medium sized
13、 firms. The Mediocredito database is a three-wave survey(19891992,19921994 and 19951997)conducted on a sample of more than 5,000 firms taken from the Italian manufacturing industry.The sample is stratified and randomly selected(it reflects the geographical and dimensional distribution of Italian fir
14、ms)for firms with 11 to 500 employees.It is organised by census for firms with more than 500 employees.In all of the three samples both qualitative and quantitative data(balance sheets for the 19891997 period)are collected.Qualitative data provide,among other things, information on ownership structu
15、re,availability of external finance,entitlement to state subsidies,and successful introduction of products and processes. Descriptive features of the sample show that more than 60 percent of the surveyed firms have less than 50 employees.Ownership structure is highly concentrated as the average shar
16、e of the first shareholder and of the control group are of about 50 percent and higher than 80 percent respectively.60 percent of firms are family-owned.In this respect the sample reflects a typical feature of many industrialised economies in which family ownership plays a relevant role. The inspect
17、ion of the quantitative variables,which are relevant for our analysis,shows other interesting features of the sample.More than 60 percent of the firms have less than 50 employees and do not invest in R&D.Return on investment is negative for 20 percent of firms and not higher than 2 percent for more
18、than 60 percent of them.More than 40 percent of the firms have only one controlling shareholder. To check whether our findings are robust with respect to the survivorship bias effect we consider the last two waves of the Survey(19921994 and 19951997).We include in the sample only those firms partici
19、pating in both surveys plus those participating in the first,but not in the second.We therefore test the effect of 1994 variables on the19951997 rate of growth. Descriptive findings on three-year average rates of growth in the last wave(19951997)for different subgroups of firms show that the overall
20、 sample three-year rate of growth is of about 4 percent for surviving firms and of 2 percent when we also consider firms which did not survive.Age and size do matter.Size accounts for a significant dispersion of this average as 30 percent of the smallest firms in the sample that survived have a 7 pe
21、rcent rate of growth against a 2 percent of the 30 percent of largest firms(a negative rate of growth of 2 percent when we correct for firms which did not survive).Using the same distribution percentiles we compare the three-year rates of growth of young and old firms,and of firms with high and low
22、levels of leverage and financial pressure.If we just consider surviving firms the rate of growth for young firms is eight times as high as that of old firms(8 percent against 1 percent).If we consider also financial variables we find that the differences in relative rates of growth among subgroups g
23、o beyond the well established stylised facts regarding size and age,already found in other countries and in different periods.Firms with higher availability of external finance(high leverage firms)grow much more than low leverage firms with the difference being more than double for firms with less t
24、han 50 employees.Firms which were credit constrained in 1994 had negative rates of growth(well below sample averages)even when not correcting for survivorship bias.Firms whose budget constraint is softened by state subsidies exhibit a relatively higher growth rate.On the contrary,firms with higher f
25、inancial pressure grow significantly less when survivorship bias is taken into account.A last strong difference is determined,as expected,by market power,with high market rent firms growing at a rate which is four times higher than the complementary subgroup The marginal contribution of this paper i
26、s that of providing a test showing that,for a representative sample of small and medium sized Italian firms,the rent-adjusted rate of growth is not due to chance and is not just affected by size and age.We find in fact that“growth independence”does not hold both in its weak and strong version(after
27、correcting for market power and industry characteristics)and that finance is not neutral in the small firm and in the overall sample also when survivorship bias is taken into account.On the contrary,in the sample of firms with more than 100 employees corrected for the survivorship bias finance is ne
28、utral and both the strong and the weak version of“growth independence”hold,confirming that the choice of the sample has critical effects on the significance of the test. Our empirical findings therefore seem to show that small surviving firms have higher than average growth potential.This potential
29、may be limited by the scarce availability of external finance and lack of access to foreign markets.These results are broadly consistent with the hypothesis that the availability of external finance and internationalisation are crucial determinants for economic development.Further empirical testing
30、on different samples and in different countries may reveal whether the bank centered financial structure of Italy has a critical role in generating the significant effects of these factors on firm growth. 译文 中小企业成长的决定因素 -外源融资的有效作用 资料来源 :EBSCO 数据库 作者 ：莱昂纳多 ,乔瓦尼 我们以意大利中小企业为样本，对其成长的决定因素做了实证分析。本文选取拥有 10
31、-50 名员工的企业作为研究样本，通过调查一些在文献里通常被认为成长能力好，预期市场竞争力强的企业，我们发现，这些企业不仅受其规模和成立时间的影响，还受到国家补助，出口能力和信用等级的影响。通过采用多种方法，在控制异方差，生存偏差，序列相关的情况下，我们的结论也被证实，我们的结论表明，大型企 业的成长一般不受到其初始规模和其他因素 （ Gib rat 定律中提到的）的影响，但是外源融资困难的中小企业会受这些因素影响。 自从 1931 年制定 Gibrat 定律，它一直是实证研究中小企业成长有效的理论标准。其中有两个主要观点可以总结如下：（ 1）在初始阶段，企业的成长与它的规模无关（ 2）在相同
32、时间段和相同产业内，成长率是相同的。第二个观点比第一个观点更加没有理论价值，它意味着在控制行业特点后，企业的成长率不受任何因素的影响。 大部分实证分析否定了企业成长与其规模和成立时间无关的假说。当只考虑幸存的企业或者考虑生存 偏差的时候，企业的成长性与它的规模和成立时间成正相关或者负相关。但是有效数据的缺乏阻止了其他方面的研究。更多详细的关于企业成长决定因素（除了企业规模和成立时间以外）的实证研究对于经济学家和政策制定者都有重要意义。举个例子，企业规模，外源融资能力，国外市场和股权结构的关系已经严重被忽视，尽管它提供了对于企业成长有重大意义的关于企业管治和融资改革的政策见解。本文旨在通过考虑一
33、系列潜在的企业成长的决定因素来填补这一空白。文章分为五部分（包括引言和结论），在第二部分，我们简要描述了理论和实证研究结果，发现对于不同国 家和不同时期，企业成长和其规模以及成立时间都有重要影响。在第三部分，我们以 1989 -1997 年之间的 4000 家意大利企业为样本找出了一些企业成长的决定因素。在第四部分，我们在控制异方差，生存偏差，序列相关的情况下得出了研究结果，表明不仅仅只有企业规模和成立时间是企业成长的决定因素，外源融资和外国市场对于中小企业也有重要影响。 中期信贷数据库是以多于 5000 家意大利制造企业为样本的三轮调查 (1989 1992, 19921994 和 1995
34、1997) 后的研究成果。样本经过分层和随机选择（反映了地域性的 意大利企业）拥有 11-500 个员工的企业。对于多于 500 个员工的企业进行统计。三个样本通过 19891997 年的企业资产负债表收集了这些企业的质量和数量数据。质量数据提供了关于企业的所有者权益结构，外源融资有效性，政府补贴以及成功产品和工艺的引进的信息。 样本的特征表明多于 60%的经过调查的企业拥有少于 50 个员工。所有者权益结构高度集中，第一股东和控制集团拥有 50%-80%的股份。 60%的企业是家族控制的。从这个角度看，样本反映 了许多产业经济的一个重要特征，家族产业扮演着相当重要的角色。从数量上观察与我们研
35、究有关的样本数据，可以发现其他有趣的特点。 60%以上的企业拥有少于 50 个员工，而且不投资 R&D。对 20%的企业投资是没有回报的，对 60%的企业投资，回报率低于 2%。 40%以上的企业只有一个股东。 为了检查我们的研究发现在生存偏差影响下是否健全，我们考虑了接下来的两轮调查 (19921994 和 19951997)。我们把参加两次调查和只参加了第一次调查的企业都包括了进来。于是我们测试了 1994 年的样本数据在 19951997年成长 率下的影响。 对不同的分组企业的第三周期 （ 19951997） 的平均成长率进行研究得出，在只考虑生存下来的企业的情况下，平均成长率为 4%，
36、如果再考虑没有生存下来的企业，平均成长率为 2%。企业规模和成立时间确实对企业成长有影响。企业规模把企业的成长率分开了，样本中 30%的生存下来的小企业的成长率为7%，其中的大企业成长率为 2%（当我们考虑那些没有生存下来的企业的时候，成长率为 -2%）。运用同样的方法，我们也比较了成立时间早和成立时间晚的企业，高负债比率和低负债比率的企业以及高融资压力和低融资压力的企业的三年的 成长率。如果我们只考虑生存下来的企业，成立时间晚的企业的成长率是成立时间早的八倍（ 8%和 1%）。如果我们也考虑融资变量，我们发现相对于企业规模和成立时间来说，融资更加影响企业的成长，这已经在其他国家和不同时期被证
37、实。对于小于 50 个员工的企业来说，有较好外源融资（高负债产权比率）企业的成长率是低负债产权比率企业的两倍。 1994 年有债务的企业在没有生存偏差因素的影响下成负增长趋势（远低于样本平均水平）。当企业的债务被国家补助缓解后，呈现了相对较高的成长率。相反的，具有融资压力的企业在考虑生存偏差后，它的成长率更 低。正如预期的一样，最后一个重要的不同就是市场因素，高市场占有率企业的成长率是其他企业的四倍。 这篇文章的作用是通过测试表明对于具有代表性的意大利的中小企业，随着租金调整的成长率不是只受其规模和成立时间的影响。我们发现事实上，“企业成长独立性”都是不成立的（在考虑市场因素和产业特征以后），在考虑生存偏差后，对于所有样本来说，外源融资不是中立的。相反的是，样本中超过 100个员工的企业纠正了以前的观点，证明了选择样本对于测试的重要影响。 因此，我们的实证研究表明生存下来的小企业比一般企业具有更高的成长潜力。这种 成长潜力也许被外源融资困难和缺少外国市场所限制。这些研究结果很大程度上符合了外源融资有效性和国际化是经济发展的决定因素。关于不同样本和不同国家的进一步的实证研究也许会揭示以银行为中心的意大利融资结构是否对这些企业成长因素起关键作用。
