经济结构决定金融结构外文文献翻译中英文最新

经济结构决定金融结构外文文献翻译中英文2019-2020
英文
Does economic structure determine financial structure?
Franklin Allen，Laura Bartiloro
Abstract
In this paper, we examine the relationship between the structure of the real economy and a country's financial system. We consider whether the development of the real economic structure can predict the direction of evolution of a country's financial structure. Using data for 108 countries, we find a significant relationship between real economic structure and financial structure. Next, we exploit shocks to the economies in India, Finland and Sweden, and South Korea and show that changes in the economic structure of a country influence the evolution of its financial system. This suggests that financial institutions and capital markets change in response to the structure of industries.
Keywords：Financial system，Economic structure
Introduction
The structures of financial systems vary among industrial and developing countries. In some countries, financial systems are predominantly bank-based, while in others they are dominated by capital markets. Only fragmented theories exist in the literature that explain the prevailing differences in country financial structures, which are defined
as the mix of financial markets, institutions, instruments, and contracts that prescribe how financial activities are organized at a particular date.
The existing studies explain the prevailing differences in financial structures using legal origin and protection, politics, history, and culture as factors. This paper considers the link between the real economic structure and the financial system of a country. Such a relationship is influenced by the funding sources for corporate investment that differ depending on firm and project characteristics (Allen, 1993; Boot and Thakor, 1997; Allen and Gale, 1999). Consistent with this theory, banks are more appropriate for the financing of traditional asset-intensive industries, whereas capital markets favour innovative and risky projects. One implication of this theory is that the real economic structure of a country, whether it is asset intensive or service oriented, could determine its financial structure. For instance, financial systems in countries such as Germany and Japan would remain bank-based as long as their economies are dominated by manufacturing industries. Contrastingly, the financial system in the United States will continue to be market-oriented as long as service and highly innovative companies constitute a large share of the economy. Consequently, the financial systems of the United States, Germany, or Japan will remain at polar extremes because of their economic structures even though the countries are at a similar stage of development.
Robinson (1952) argues that financial intermediaries and markets emerge when required by industries. Consequently, intermediaries and markets appear in response to economic structure. The idea that the form of financing, and thus the country's financial structure, depends on the type of activity that firms engage in has not yet been directly addressed in the literature. To provide evidence of the hypothesis that structure and changes in the real economy determine the direction of evolution of a country's financial system, we first must distinguish the different financial structures across countries. However, although recent attention has shifted to a more systematic classification of financial systems, the literature provides only very broad measures and definitions for classification. Consistent with the literature this study classifies a country's financial system as either bank-based (the German or Japanese model) or market-based (the Anglo-Saxon model). In the bank-based financial system, financial intermediaries play an important role by mobilising savings, allocating credit, and facilitating the hedging, pooling, and pricing of risks. In the market-based financial system, capital markets are the main channels of finance in the economy (Allen and Gale, 2000).
Our theory builds on Rajan and Zingales (2003a) who note that bank-based systems tend to have a comparative advantage in financing fixed-asset-intensive firms rather than high technology research and development-based firms. Rajan and Zingales (2003a) argue that
fixed-asset-intensive firms are typically more traditional and well understood, and the borrower has the collateral to entice fresh lenders if the existing ones prove overly demanding. As per Rajan and Zingales (2003a), loans are well collateralised by physical assets, and therefore are liquid; hence, the concentration of information in the system will not be a barrier to the financing of these assets. Conversely, the authors argue that market-based systems will have a comparative advantage in financing knowledge industries with intangible assets.
Consequently, we suggest that countries with a majority of physical-asset-intensive firms, depending on external finance, will be more likely to possess a bank-oriented financial system. However, capital markets should develop more effectively in countries with firms that are based on knowledge and intangible assets. We test this hypothesis by identifying fixed-asset-intensive firms within the economic sector defined as industry by the standard classification system for economic activity. Conversely, in this study the service sector acts as a proxy for knowledge and intangible asset firms. The relative importance of the two types of firms in an economy will be represented by the relative volume of activity of the two different economic sectors. The standard system of classification for economic activity includes a third sector, agriculture. We classify agriculture as a physical-asset-intensive industry because land and agricultural machinery may be used as collateral and, therefore,
we assume that firms in the agricultural sector will prefer bank financing over capital markets.
We first present some historical evidence showing the nexus between real economic structure and financial system. In order to test our outlined hypothesis, we use a panel data set for 108 countries and employ both the panel OLS and a two-step generalised-method-of-moments (GMM) system. Additionally, we investigate the robustness of the results by introducing different additional control variables and testing the heterogenous effects. The results suggest that there is a negative and significant relationship between a country's economic structure (industry versus service sector) and financial system structure (stock market versus banking sector). In economies where the service sector carries more weight economically than industry and agriculture, the country tends to have a market-based financial system. In contrast, a bank-based financial system is more likely to emerge in economies with many fixed-asset-intensive firms.
Next, we conduct event studies using the treatment effect estimation to isolate the endogeneity concerns. We analyse different types of exogenous shock to the structure of the real economy and its impact on financial structure. We employ three events that changed the economic structures of the countries and further investigate their impact on the financial structure using a difference-in-difference strategy. The first
event is India's structural reforms in 1991 as a positive shock to the country's economy; the second one is the demise of the Soviet Union as a negative shock to the economy of Finland and Sweden; the third one is the economic reforms in the 1980s and early 1990s in South Korea. In India and South Korea we find that after the structural reforms of the economies, the service sector grew in relative terms and the stock markets in both countries experienced significantly faster growth than their banking systems, compared to the control countries. In Finland and Sweden we document that following the negative shock the service sector gained in relative importance, which was followed by the faster growth of the equity market in comparison to the banking system. Overall, the results of the three different event studies confirm our hypothesis that the relative importance of financial intermediaries and markets is determined by the industry needs of a country.
The findings of this study are interesting from a regulatory perspective and lend insight into the development of financial structures worldwide. The main policy implications from this study are that financial structures should be evaluated in terms of whether they meet the requirements of the real economy and industries. Furthermore the financial structure cannot be changed as long as the economic structure does not change. The results provide insight into the reasons for limited capital markets growth in developing countries despite official
stimulation efforts from governments and multilateral organisations (Schmukler et al., 2007). According to our study of many developing countries, as long as economies remain relatively agriculture- and industry-oriented, any government effort to create or further develop a capital market is likely to not to be very successful. Additionally, any regulation that attempts to force a change in the financial system may result in a discrepancy in the economic and financial structure. Therefore, such efforts or regulations may introduce financial constraints that can further stall economic growth because financial structure influences output levels and economic growth (Levine and Zervos, 1998; Luintel et al., 2008).
The real economy and finance nexus
A number of explanations for financial structure exist in the literature; however, none are able to provide a comprehensive account of the observations. The first explanation is based on legal origin and investor protection. Levine (1997) builds on the work of La Porta et al. (1997, 1998); henceforth LLSV) stating that legal systems originate from a limited number of legal traditions: English common law or French, German, and Scandinavian civil law. In his study on financial development and economic growth, the author employs measures of creditors' rights and demonstrates that they may explain the emergence of bank-based financial systems. Modigliani and Perotti (2000) argue
that legal institutionsdetermine the degree of financial development and the financial structure of a country. They argue that market-based systems flourish in environments with strong institutions. Ergungor (2004) also attempts to explain differences in financial structure by examining legal origin across countries. His study presents evidence that countries with civil law financial systems are more likely to be bank-oriented than common law countries. In the author's opinion, this evolution is a result of effective rule of law in common law countries, which improves shareholder and creditor rights protection. A perspective has emerged in the literature that legal origin can be used to explain the structure of a financial system.
However, Rajan and Zingales (2003a) argue that countries with a common law system did not rely on markets to a greater extent than civil law systems at the beginning of the last century. They report that in 1913, the ratio of France's stock market capitalisation to GDP was twice as high as that of the United States, which is a country that has an environment that favours capital market development according to the legal origin perspective. It is therefore problematic to argue that legal origin is the main determinant of financial structure. The view presented below is that both the structure of the financial system and the laws will adapt to the needs and demands of the economy. One example of this is branching regulation in the United States banking sector. Rajan and Zingales
(2004) note that as technology improved the ability of banks to lend and borrow from customers at a distance, competition increased in the United States even when banks had no in-state branches. Politicians who could not prevent this competition because they lacked jurisdiction, withdrew the regulations that limited branching. Another example is the removal of the Glass-Steagall Act, which had restricted banking activities in the United States since 1933. In this case, the introduction of the Financial Modernisation Act in 1999 followed the creation of the first financial holding company in the United States and removed past restrictions. Therefore, we argue that economic demand may enhance the evolution of the financial structures and of the legal system.
The existing empirical results show also that legal investor protection may support financial development. For example, LLSV (1997) show that countries with poorer investor protection have less developed capital markets. Demirgüç-Kunt and Levine (2004) find that countries with stronger protection for shareholder rights tend to have a more market-based financial system. Djankov et al. (2007) investigate cross-country determinants of private credit and find that legal creditor rights are statistically significant and quantitatively important in determining private credit development, while there is no evidence showing that creditor rights are converging among legal origins. Moreover, Djankov et al. (2007) confirm that shareholder protection is
positively related to stock market development.
The second explanation for financial structure is based on political factors. Biais and Perotti (2002) provide a theoretical model of government incentives to structure privatisation policy so that financial shareholders are diffused, which may be designed to ensure re-election. Additionally, Perotti and V olpin (2004)argue that established firms have an incentive to limit entry by retarding financial development, which may well impact the financial structure. Perotti and von Thadden (2006) use a theoretical model to demonstrate the effect of the distribution of income and wealth in democratic societies and their influence on the financial structure of an economy.
Moreover, according to Rajan and Zingales (2003a, 2004) structures of the financial system are unstable and evolve over time. They argue that a financial system will develop toward the optimal structure but will be hindered by politics, which are often influenced by powerful, incumbent groups. Similarly, Cull and Xu (2013) argue that financial development is driven by political economy. In their opinion financial development may reflect the interests of the elite, rather than providing broad-based access to financial services. Song and Thakor (2012a, b) develop a theory of how a financial system is influenced by political intervention that is designed to expand credit availability. They show that the relationship between political intervention and financial system development is
nonmonotonic. In the early stage of financial development, the size of markets is relatively small and politicians intervene by controlling some banks and providing capital subsidies, while in the advanced stage when the financial sector is most developed, political intervention returns in the form of direct-lending regulation.
Industrial-level evidence
To check the robustness of our main results we conduct a wide array of additional analyses; however, for brevity we do not report them in full. First, we check the consistency of the results after removing outliers. These outliers are eliminated after considering the scatter plot of the main financial and economic structure indicators. We eliminate those countries that fall particularly far from the regression line and then repeat the estimation on the new sample. After eliminating the extreme observations, we still find a significant and negative relationship between economic and financial structure. Second, we increase the set of explanatory variables and add variables for country GDP, inflation, area, latitude, dummies for landlocked economies, transition economies, or developing countries. Including these variables does not affect either the significance level or the sign of the estimated coefficients. Third, we divide the countries in the sample into two groups based on their membership in the OECD. We assume that countries belonging to the OECD are on average more developed than non-OECD member countries. Using the two separate
samples we compute again the baseline regressions. The results indicate that the relationship between financial structure and economic structure is much stronger in industrial countries than in developing countries. One possible explanation for this result is the different development stage of the financial system itself. In developing countries, the financial structure is emerging and adjusting to the needs of the real economy at the same time. Moreover, rapid changes in the financial structure are often caused by additional factors such as liberalization or political transformation. Conversely, in most of the industrial countries, we may assume that the financial system may already have an optimal structure, whereas changes are only caused in case of significant changes in the economic structure, which takes substantial time.
Fourth, in the case of the OECD countries the data availability on the composition of value added for most of the industries allows us to calculate an alternative measure of economic structures, where we control for the firm asset characteristics in the given industry. In this analysis, the primary data source is the OECD STAN database for industrial analysis, which enables retrieval of gross value added for 47 industries representing nine main sectors of the economy in 25 countries. We divide the industries using firm specific characteristics from either an asset-intensive or knowledge sector, where we measured asset intensity as the ratio of tangible assets (property, plant and equipment) to total book
assets of the firm in the industry, whereas the company specific data was computed using data from the Bureau van Dijk's ORBIS database.
According to our theory, asset-intensive firms with tangible assets may use the assets to collateralise their bank debt. Hence, in countries dominated by asset-intensive industries bank-based financial systems are more likely to emerge. In contrast, knowledge-based companies with a low level of tangible assets are often forced to use either equity or bonds to finance their needs. Therefore, countries dominated by industries with intangible assets are more likely to have a market-based financial system.
Classifying industries as either asset or tangible asset intensive, where we distinguish industries using ratios calculated on firm level data, we again construct two alternative measures for economic structure and employ them in the basic regression. The results of those regressions are similar to those we have presented previously and the coefficients of the economic structure were again negative and statistically significant. Overall the robustness tests at the industry-level also confirm our findings on the link between economic structure and financial structure.
Conclusion
Our results provide new evidence concerning the causes and causality of the direction of evolution of the financial system structure. Using both OLS, dynamic panel techniques and event studies we
document that the economic structure is closely linked to the shape of the financial system. We find that countries with asset-intensive sectors are more likely to have a bank-based system. Conversely, countries with sectors that are based on knowledge and intangible assets are likely to exhibit a market-based financial system. The results suggest that the structure of the real economy may influence the structure of the financial system. Additionally, even during systemic crises, such a relationship still holds. Moreover, we conduct event studies using a difference-in-difference strategy in order to address the problems of potential endogeneity. We use different shocks that alter the economic structures in India, Finland and Sweden, and South Korea. In all the countries the shocks resulted in significant development of the service sector relative to the industry sector. The changes in economic structure were followed by changes in the structure of financial system, where the stock market gained on importance relative to the banking sector. Consequently, the results of the event studies suggest a causal relation between economic structure and financial structure.
In our opinion, these results present a missing link in the explanation as to why country financial structures still differ. The results, however, confirm that other factors may influence the structure of the financial system. Consequently, a financial system may not always have an optimal structure, which may be a result of political arrangements or the interests
of incumbent groups (Rajan and Zingales, 2003a,b). Therefore, we assume that financial systems may not always be able to reach their optimal structure. However, as existing barriers are removed the structure of a financial system may develop and gain ground, but it would be independent of further changes in the real economic structure. Finally, when the financial system has reached its optimal structure with respect to the characteristics of the real economy, our theory implies that any increase in the significance of fixed-asset-intensive sectors would lead to an increase in the role of banks with respect to the stock market.
The main policy implications of the model are that despite efforts from governments and multilateral organisations, particular those from the emerging economies, country capital markets will not grow in size or activity as long as the economy remains asset-intensive. Therefore, governments should focus on improving the transparency or efficiency of the existing financial structure and less on the development of the stock market because the market will develop as soon as the economic structure changes. These results are consistent with Robinson (1952).
Finally, this study contributes to the ongoing debate on the relative merits of bank-based versus market-based financial systems with respect to the promotion of economic growth. Our paper presents plausible explanations to Luintel et al. (2008), that financial structure matters with respect to economic growth.
中文
经济结构决定金融结构吗？
富兰克林·艾伦，劳拉·巴蒂洛罗
摘要
在本文中，我们研究了实体经济结构与一国金融体系之间的关系。

我们考虑实体经济结构的发展是否可以预测一个国家金融结构的发展方向。

使用108个国家/地区的数据，我们发现实际经济结构与金融结构之间存在显着的关系。

接下来，我们利用对印度，芬兰，瑞典和韩国经济的冲击，并表明一国经济结构的变化会影响其金融体系的发展。

这表明金融机构和资本市场是根据产业结构而变化的。

关键词：金融体系，经济结构
引言
工业国家和发展中国家的金融体系结构各不相同。

在某些国家，金融体系主要以银行为基础，而在另一些国家，则以资本市场为主导。

文献中仅存在零散的理论，这些理论解释了国家金融结构中的普遍差异，这些差异定义为规定了特定日期的金融活动组织方式的金融市场，机构，工具和合同的组合。

现有研究使用法律渊源和保护，政治，历史和文化作为因素来解释金融结构中的普遍差异。

本文考虑了一个国家的实际经济结构与金融体系之间的联系。

这种关系受到公司投资资金来源的影响，资金来源根据公司和项目的特点而有所不同（Allen，1993；Boot和Thakor，1997；Allen和Gale，1999）。

与该理论一致，银行更适合于传统的
资产密集型产业的融资，而资本市场则偏向创新和高风险的项目。

该理论的一个含义是，一个国家的实际经济结构，无论是资产密集型还是服务型，都可以决定其金融结构。

例如，只要德国和日本等国家的经济以制造业为主导，它们的金融体系将仍然以银行为基础。

相反，只要服务和高度创新的公司在经济中占很大比重，美国的金融体系将继续以市场为导向。

因此，即使美国，德国或日本的经济结构处于相似的发展阶段，它们的经济结构仍将处于极端的极端状态。

罗宾逊（Robinson，1952）认为，金融中介和市场是在行业需要时出现的。

因此，中介机构和市场是对经济结构的回应。

关于融资形式以及国家金融结构取决于企业从事的活动类型的想法尚未在文献中得到直接解决。

为了提供有关实体经济的结构和变化决定一个国家金融体系发展方向的假说的证据，我们首先必须区分国家之间的不同金融结构。

但是，尽管最近的注意力已转移到对金融系统进行更系统的分类，但文献仅提供了非常广泛的分类方法和定义。

与文献相一致，本研究将一个国家的金融系统分为基于银行的模型（德国或日本模型）或基于市场的模型（盎格鲁-撒克逊模型）。

在基于银行的金融系统中，金融中介机构通过调动储蓄，分配信贷以及促进对冲，分担和定价风险发挥了重要作用。

在基于市场的金融体系中，资本市场是经济中金融的主要渠道（Allen and Gale，2000）。

我们的理论是建立在Rajan和Zingales（2003a）的基础上的，他们指出，基于银行的系统在为固定资产密集型企业提供融资方面比在高科技研发为基础的公司上具有比较优势。

Rajan和Zingales（2003a）
认为，固定资产密集型公司通常更为传统，而且也被很好地理解，如果现有贷款人的需求过高，借款人便可以吸引新的贷款人。

根据Rajan 和Zingales（2003a），贷款以有形资产作抵押，因此具有流动性。

因此，信息在系统中的集中不会成为这些资产融资的障碍。

相反，作者认为，基于市场的系统在利用无形资产为知识产业融资方面将具有比较优势。

因此，我们建议，那些拥有大量实体资产密集型公司的国家（取决于外部金融）将更有可能拥有面向银行的金融体系。

但是，在拥有以知识和无形资产为基础的公司的国家，资本市场应该更有效地发展。

我们通过确定经济活动的标准分类系统将其定义为行业的经济部门中的固定资产密集型公司来检验该假设。

相反，在这项研究中，服务业是知识和无形资产公司的代理。

两种类型的公司在经济中的相对重要性将由两种不同经济部门的相对活动量来表示。

经济活动的标准分类系统包括第三部门，即农业。

由于土地和农业机械可能被用作抵押品，因此我们将农业归类为物质密集型产业，因此，我们假设农业部门的公司比资本市场更倾向于银行融资。

我们首先提出一些历史证据，以显示实际经济结构与金融体系之间的联系。

为了检验概述的假设，我们使用了108个国家/地区的面板数据集，并采用了面板OLS和两步广义矩量法（GMM）系统。

此外，我们通过引入不同的其他控制变量并测试异质效应来研究结果的鲁棒性。

结果表明，一国的经济结构（工业与服务业）与金融体系结构（股票市场与银行业）之间存在负相关关系。

在服务业在经济上比
工业和农业更重要的经济体中，该国倾向于建立以市场为基础的金融体系。

相反，在拥有许多固定资产密集型公司的经济体中，更可能出现基于银行的金融体系。

接下来，我们使用治疗效果估算进行事件研究，以隔离内生性问题。

我们分析了对实体经济结构及其对金融结构的影响的不同类型的外源冲击。

我们采用了三种改变国家经济结构的事件，并使用差异策略进一步研究了它们对金融结构的影响。

首先是印度在1991年进行了结构性改革，这对该国的经济产生了积极的冲击。

第二个因素是苏联的消亡，这对芬兰和瑞典的经济造成了负面冲击。

第三个是1980年代和1990年代初期韩国的经济改革。

在印度和韩国，我们发现，在经济体制改革之后，服务业相对增长，与控制国相比，两国的股票市场增长速度都快于其银行系统。

在芬兰和瑞典，我们记录到，在遭受负面冲击之后，服务业变得相对重要，其次是股票市场比银行体系更快的增长。

总体而言，三项不同事件研究的结果证实了我们的假设，即金融中介机构和市场的相对重要性取决于一个国家的行业需求。

从监管的角度来看，这项研究的结果是有趣的，并有助于洞察全球金融结构的发展。

这项研究的主要政策含义是，应根据金融结构是否满足实体经济和产业的要求对其进行评估。

此外，只要经济结构不改变，金融结构就不能改变。

尽管政府和多边组织作出了官方的刺激努力，但研究结果提供了对发展中国家资本市场增长有限的原因的见解（Schmukler等，2007）。

根据我们对许多发展中国家的研究，只要经济仍然以农业和工业为导向，那么政府为建立或进一步发展资本市
场所做的任何努力都可能不会很成功。

此外，任何试图迫使金融体系发生变化的法规都可能导致经济和金融结构出现差异。

因此，由于金融结构影响产出水平和经济增长，此类努力或法规可能会引入金融约束，从而进一步阻碍经济增长（Levine和Zervos，1998；Luintel等，2008）。

实体经济与金融的联系
文献中对金融结构有许多解释；但是，没有人能够全面介绍这些观察结果。

第一种解释是基于法律渊源和投资者保护。

Levine（1997）建立在La Porta等人（1997，1998）的工作的基础上。

此后LLSV指出，法律制度源自有限的法律传统：英国普通法或法国，德国和斯堪的纳维亚民法。

在他关于金融发展和经济增长的研究中，作者采用了债权的衡量标准，并证明它们可以解释基于银行的金融体系的出现。

Modigliani和Perotti（2000）认为，法律制度决定了一个国家的金融发展程度和金融结构。

他们认为，基于市场的系统在具有强大制度的环境中会蓬勃发展。

Ergungor（2004）还试图通过研究各国的法律渊源来解释金融结构的差异。

他的研究表明，拥有民法金融体系的国家比普通法国家更有可能面向银行。

作者认为，这种演变是普通法国家有效法治的结果，可以改善股东和债权人的权利保护。

文献中出现了一种观点，即合法来源可以用来解释金融体系的结构。

但是，拉詹和津加莱斯（Rajan and Zingales，2003a）认为，上世纪初，拥有习惯法体系的国家比民法体系更不依赖市场。

他们报告说，在1913年，法国的股票市值对GDP的比率是美国的两倍，而美国的
法律根据法律渊源的观点，其环境有利于资本市场的发展。

因此，争论法律渊源是金融结构的主要决定因素是有问题的。

下面的观点是，金融体系的结构和法律都将适应经济的需求。

其中一个例子是美国银行业的分支机构监管。

Rajan和Zingales（2004）指出，随着技术的发展，银行提高了远距离向客户借贷的能力，即使在美国银行没有州内分行的情况下，竞争也在加剧。

由于缺乏管辖权而无法阻止这场竞争的政客撤回了限制分支机构的规定。

另一个例子是取消了《格拉斯－斯蒂格尔法案》，该法案自1933年以来一直限制了美国的银行业务。

在这种情况下，1999年引入《金融现代化法案》之后，在美国成立了第一家金融控股公司。

并删除了过去的限制。

因此，我们认为经济需求可能会促进金融结构和法律体系的发展。

现有的经验结果还表明，合法的投资者保护可以支持金融发展。

例如，LLSV（1997）表明，投资者保护较差的国家资本市场不发达。

Demirgüç-Kunt和Levine（2004）发现，对股东权益有更强保护的国家往往拥有更加以市场为基础的金融体系。

Djankov （2007）调查跨国私人信贷的决定因素，发现合法的债权在决定私人信贷发展方面具有统计上的重要性和数量上的重要性，而没有证据表明债权在法律渊源之间趋于一致。

此外，Djankov等（2007年）证实，股东保护与股票市场发展成正相关。

对金融结构的第二种解释是基于政治因素。

Biais和Perotti （2002）提供了一种政府激励机制的理论模型，以构建私有化政策，从而分散了财务股东的利益，这可以用来确保连续性。

此外，Perotti
和V olpin（2004）认为成立公司有动机通过限制金融发展来限制进入，这很可能会影响金融结构。

Perotti和von Thadden（2006）使用理论模型来证明民主社会中收入和财富分配的影响及其对经济金融结构的影响。

而且，根据Rajan和Zingales（2003a，2004）的观点，金融体系的结构是不稳定的，并且会随着时间而变化。

他们认为，金融体系将朝着最佳结构发展，但会受到政治的阻碍，而政治通常会受到强大的在位集团的影响。

同样，Cull和Xu（2013）认为金融发展是由政治经济学驱动的。

他们认为，金融发展可能反映了精英阶层的利益，而不是提供广泛的金融服务渠道。

Song and Thakor（2012a，b）提出了一种理论，该理论旨在通过旨在扩大信贷可用性的政治干预来影响金融体系。

他们表明，政治干预与金融体系发展之间的关系是非单调的。

在金融发展的初期，市场规模相对较小，政客通过控制一些银行并提供资本补贴进行干预。

证据
为了检查主要结果的稳健性，我们进行了大量其他分析。

但是，为了简洁起见，我们不会将其完整报告。

首先，我们在去除异常值后检查结果的一致性。

在考虑了主要金融和经济结构指标的散点图之后，消除了这些离群值。

我们剔除了那些离回归线特别远的国家，然后对新样本重复进行估算。

在消除极端观察结果之后，我们仍然发现经济和金融结构之间存在显着的负面关系。

其次，我们增加了解释变量集，并增加了国家GDP，通货膨胀，面积，纬度，内陆经济体，
转型经济体或发展中国家的虚拟变量的变量。

包括这些变量不会影响显着性水平或估计系数的符号。

第三，我们根据样本国家在经合组织中的成员身份将其分为两组。

我们假设属于经合组织的国家平均比非经合组织成员国更发达。

使用两个单独的样本，我们再次计算基线回归。

结果表明，工业化国家的金融结构与经济结构之间的关系要强于发展中国家。

对这一结果的一种可能解释是金融体系本身的发展阶段不同。

在发展中国家，金融结构正在出现，并同时适应实体经济的需求。

此外，金融结构的快速变化通常是由自由化或政治转型等其他因素引起的。

相反，在大多数工业国家中，我们可以假设金融体系可能已经具有最佳结构，而变化仅是在经济结构发生重大变化的情况下才引起的，这需要花费大量时间。

第四，在经合组织国家中，大多数行业增加值构成的数据可用性使我们能够计算经济结构的替代指标，在此基础上我们可以控制给定行业中公司资产的特征。

在此分析中，主要数据源是用于工业分析的OECD STAN数据库，该数据库可检索代表25个国家/地区的9个主要经济部门的47个行业的总增加值。

我们将利用公司特定特征的行业从资产密集型或知识部门中划分出来，在该部门中，我们以资产有形资产（物业，厂房和设备）与该行业中公司账面总资产的比率来衡量资产强度。

特定数据是使用Bureau van Dijk的ORBIS数据库中的数据计算得出的。

根据我们的理论，拥有有形资产的资产密集型企业可能会使用这些资产抵押其银行债务。

因此，在以资产密集型产业为主导的国家中，。