Journal Issue

Introducing a New Broad-based Index of Financial Development1

Katsiaryna Svirydzenka
Published Date:
January 2016
  • ShareShare

Related Material

Show Summary Details

I. Introduction

A large body of literature has developed to assess the impact of financial development on economic growth, inequality, and economic stability (see Levine, 2005, Demirgüç-Kunt and Levine, 2009, and Dabla-Norris and Srivisal, 2013 for respective literature surveys). Financial development involves improvements in such functions provided by the financial systems as: (i) pooling of savings; (ii) allocating capital to productive investments; (iii) monitoring those investments; (iv) risk diversification; and (v) exchange of goods and services (Levine, 2005). Each of these financial functions can influence saving and investment decisions and the efficiency with which funds are allocated. As a result, finance affects the accumulation of physical and human capital and total factor productivity – the three factors that determine economic growth. To the extent that financial development reduces informational asymmetries and financial constraints and promotes risk sharing, it can enhance the ability of financial systems to absorb shocks and reduce the amplification of cycles through the financial accelerator (Bernanke, Gertler, and Gilchrist 1999), lowering macroeconomic volatility and inequality.

Most of the empirical literature since the 1970s approximates financial development by two measures of financial depth – the ratio of private credit to GDP and, to a lesser extent, by stock market capitalization, also as a ratio to GDP. For example, in an influential industry-level study Rajan and Zingales (1998) use both measures to show that more financial development facilitates economic growth. More recently, Arcand, Berkes, and Panizza (2012) use credit to GDP ratio to establish that there is a threshold above which financial development no longer has a positive effect on economic growth. On the macroeconomic volatility side, Dabla-Norris and Srivisal (2013) find that financial development, as measured by private credit to GDP from banks and other financial institutions, plays a significant role in dampening the volatility of output, consumption, and investment growth, but only up to a certain point. Most researchers in this field use variations of these two measures to examine the role of the financial system in economic development.

And yet, financial development is a multidimensional process. With the passage of time, financial sectors have evolved across the globe and modern financial systems have become multifaceted. For example, while banks are typically the largest and most important, investment banks, insurance companies, mutual funds, pension funds, venture capital firms, and many other types of nonbank financial institutions now play substantive roles. Similarly, financial markets have developed in ways that allow individuals and firms to diversify their savings, and firms can now raise money through stocks, bonds, and wholesale money markets, by-passing traditional bank lending. The constellation of such financial institutions and markets facilitates the provision of financial services. Furthermore, an important feature of financial systems is their access and efficiency. Large financial systems are of limited use if they are not accessible to a sufficiently large proportion of the population and firms. Even if financial systems are sizeable and have a broad reach, their contribution to economic development would be limited if they were wasteful and inefficient. This point is made also, for example, in Čihák et al. (2012) and Aizenman, Jinjarak, and Park (2015). The diversity of financial systems across countries implies that one needs to look at multiple indicators to measure financial development.

To overcome the shortcomings of single indicators as proxies for financial development, we create a number of indices that summarize how developed financial institutions and financial markets are in terms of their depth, access, and efficiency, culminating in the final index of financial development (Figure 1). These indices were originally developed in the context of the IMF Staff Discussion Note “Rethinking Financial Deepening: Stability and Growth in Emerging Markets” (Sahay et al., 2015). This paper presents and explains the methodology that underpins them. The sub-indices and the final overall index are constructed for 183 countries on annual frequency between 1980 and 2013. Financial institutions include banks, insurance companies, mutual funds, and pension funds. Financial markets include stock and bond markets. Financial development is defined as a combination of depth (size and liquidity of markets), access (ability of individuals and companies to access financial services), and efficiency (ability of institutions to provide financial services at low cost and with sustainable revenues, and the level of activity of capital markets). This broad multi-dimensional approach to defining financial development follows the matrix of financial system characteristics developed by Čihák et al. (2012).

Figure 1.Financial Development Index Pyramid

Source: IMF staff, based on Čihák and et al. (2012)

While this paper follows Čihák et al. (2012) in their conceptual approach to defining financial development, the contribution of this paper is quite distinct. First, we supplement the World Bank FinStats, a more updated version of the Global Financial Development Database (GFDD) introduced by Čihák and co-authors, with additional data from the Bank of International Settlements (BIS) debt securities database, Dealogic corporate debt database, and IMF Financial Access Survey. Second, we summarize this diverse information in several easy to use indices. Given the wealth of information on financial system properties – there are 105 distinct indicators in GFDD and 46 indicators in FinStats – it is not feasible to track all of these different indicators individually, especially in empirical work. And even if it were possible, not one single indicator, when taken on its own, would offer a comprehensive understanding of the level of financial development. The sub-indices and the final index pull together these various indicators and allow a comprehensive assessment of particular features of financial systems and the overall level of financial development. As a result, the indices allow to pin down where deficiencies in financial development lie or which aspects of financial development affect macroeconomic performance, which could then be investigated in greater detail using the disaggregated data from FinStats or GFDD.

The paper also provides additional robustness checks and deviates in some ways from the Staff Discussion Note. More specifically, the number of issuers (one of the indicators for financial market access) is now scaled by population to bring it in line with financial institutions access measures. As a result, the relative ranking of some countries changes, with countries with larger population receiving lower scores in the new database compared to the original release. Missing data treatment is now applied iteratively and no longer uses the data on profit growth in reconstruction. The database is also updated for the more recent releases of Finstats and Dealogic data.

In what follows, the paper describes the methodology used to construct the indices, including data sources, treatment of missing values, functional form and weights used in aggregation. It shows how the new indices compare with traditional measures and the key stylized facts on financial development across the globe. It concludes with a discussion of some caveats and limitations of the indices.

II. Methodology

The financial development index is constructed using a standard three-step approach found in the literature on reducing multidimensional data into one summary index: (i) normalization of variables; (ii) aggregation of normalized variables into the sub-indices representing a particular functional dimension; and (iii) aggregation of the sub-indices into the final index. This procedure follows the OECD Handbook on Constructing Composite Indicators (OECD, 2008), which is a good reference for methodological suggestions. There are a number of examples in the literature of constructing composite indices that compare and rank country performance. These include the IMF Financial Stress Index (Cardarelli, Elekdag, and Lall, 2008; Cardarelli, Elekdag, and Lall, 2009), various financial inclusion indices (Amidžić, Massara, and Mialou, 2014; Camara and Tuesta, 2014), and the United Nations Development Programme well-being indices, such as the Human Development Index, Gender-Inequality Index, Gender Development Index, and Multidimensional Poverty Index (UNDP, 2014).

For this paper, we construct a total of nine indices, which assess at varying levels of abstraction how developed financial systems are across countries. Starting from the bottom of the pyramid in Figure 1, six lower level sub-indices are constructed using a list of indicators to measure how deep, accessible, and efficient financial institutions and financial markets are. These sub-indices are called FID, FIA, FIE, FMD, FMA, and FME, where the letters I and M denote institutions and markets, and the letters D, A, and E denote depth, access, and efficiency. These sub-indices are aggregated into two higher level sub-indices, FI and FM, which measure how developed financial institutions and financial markets are overall. Finally, FI and FM sub-indices are aggregated into the overall measure of financial development – the FD index.

A number of choices need to be made in the process of the index construction: (i) which data series to use for the sub-indices; (ii) how to treat missing data; and (iii) normalization and treatment of outliers; (iv) functional form of the aggregator; and (v) weights to use in the aggregation. These choices are covered in the subsections below.

A. Data sources

The dataset puts together 33 years of annual data between 1980 and 2013 for 183 advanced, emerging, and low-income developing countries. It draws on a number of data sources: the World Bank FinStats 2015 (Feyen, Kibuuka, and Sourrouille, 2014), IMF’s Financial Access Survey, Dealogic corporate debt database, and Bank for International Settlement (BIS) debt securities database.

A set of key indicators is chosen to capture the different aspects of the financial system characteristics (Table 1). Only variables that cover a sufficiently wide range of countries across a sufficiently long time period are selected. As a result, a number of potentially useful indicators could not be included, as discussed below. Instead, the database relies on a set of key proxy variables that may have limitations, but are well established and available for a broad country-time sample. Table 1 provides detailed information on data sources, and Table 2 gives the summary statistics of the raw data.

Table 1.Data Sources
Financial Institutions
DepthPrivate-sector credit to GDPFinStats 2015
Pension fund assets to GDPFinStats 2015
Mutual fund assets to GDPFinStats 2015
Insurance premiums, life and non-life to GDPFinStats 2015
AccessBank branches per 100,000 adultsFinStats 2015
ATMs per 100,000 adultsIMF Financial Access Survey
EfficiencyNet interest marginFinStats 2015
Lending-deposits spreadFinStats 2015
Non-interest income to total incomeFinStats 2015
Overhead costs to total assetsFinStats 2015
Return on assetsFinStats 2015
Return on equityFinStats 2015
Financial Markets
DepthStock market capitalization to GDPFinStats 2015
Stocks traded to GDPFinStats 2015
International debt securities of government to GDPBIS debt securities database
Total debt securities of financial corporations to GDPDealogic corporate debt database
Total debt securities of nonfinancial corporations to GDPDealogic corporate debt database
AccessPercent of market capitalization outside of top 10 largest companiesFinStats 2015
Total number of issuers of debt (domestic and external, nonfinancial and financial corporations)FinStats 2015
EfficiencyStock market turnover ratio (stocks traded to capitalization)FinStats 2015
Source: IMF staff estimates.
Source: IMF staff estimates.
Table 2.Summary Statistics of the Underlying Data
CodeNameObsMeanMedian St.Dev.MinMax
Financial Institutions Depth
FID1Private sector credit to GDP5,3284330390.30319
FID2Pension fund assets to GDP942208280.00157
FID3Mutual fund assets to GDP97287105190.005,232
FID4Insurance premiums (life + non-life) to GDP3,3713230.0118
Financial Institutions Access
FIA1Bank branches per 100,000 adults1,7221813180.1398
FIA2ATMs per 100,000 adults1,5164028430.01290
Financial Institutions Efficiency
FIE1Net interest margin3,3915440.0244
FIE2Lending-deposits spread4,7508680.0392
FIE3Non-interest income to total income3,5273937160.01100
FIE4Overhead costs to total assets3,4194330.0448
FIE5Return on assets3,434113−10921
FIE6Return on equity3,422121445−1,792192
Financial Markets Depth
FMD1Stock market capitalization to GDP2,5174526570.00549
FMD2Stocks traded to GDP2,312285580.000756
FMD3International debt securities of government to GDP1,56484100.00398
FMD4Total debt securities of financial corporation to GDP1,7512531030.0001,912
FMD5Total debt securities of nonfinancial corporation to GDP2,229156250.000341
Financial Markets Access
FMA1Percent of market capitalization outside of top 10 largest companies6695553191499
FMA2Total number of issuers of debt (domestic and external, fin. and non-fin. corporations) per 100,000 adults1,8040.30.10.608
Financial Markets Efficiency
FME1Stock market turnover ratio (value traded/stock market capitalization)2,3134322570.01581
Source: IMF staff estimates.
Source: IMF staff estimates.

For a small number of countries, private sector credit data is adjusted for butt splicing. More specifically, the data is corrected for one off jumps in the coverage of the banking system by the International Financial Statistics (IFS) database, the original source of FinStats data. This is only the case for a few advanced countries, when credit to GDP more than doubles in a single year as a result of revisions in banking system coverage. For example, in original source data, credit to GDP in Denmark jumps from 30 percent of GDP in 1999 to 135 percent in 2000, and in Sweden it jumps from 40 to 93 percent of GDP in 2001. If taken at face value, it would imply an impressive increase in financial depth, rather than data revisions. The data are corrected to take the most recent level as the most representative and downward shifts are merged through growth rates. This adjustment is cross checked with IFS notes on data breaks and does not affect the gradual buildup in credit during credit booms (Thailand in late 1990s, Cyprus and Iceland in 2000s) or jumps in credit in crisis or hyperinflation episodes (Argentina and Brazil at the end of 1980s and in early 1990s).

Financial institutions depth sub-index then adds to the standard banking sector depth measure used in the literature (bank credit to the private sector) indicators for other financial institutions: the assets of the mutual fund and pension fund industries and the size of life and non-life insurance premiums. Insurance premiums data is preferred to insurance companies assets because it covers more countries (a maximum of 153 versus 128) for a longer time period (starting in 1990 as opposed to 2000). As a result the country-year coverage doubles for this indicator.

Financial institution access and efficiency measures are more bank specific, given the lack of this information for other financial institutions. Financial institutions access is proxied by the number of bank branches and ATMs per 100,000 adults. Additional indicators were considered, such as the number of bank accounts per 1,000 adults, percent of firms with line of credit, and usage of mobile phones to send and receive money. These indicators were not included in the sub-index because they lack sufficiently large country and time coverage. For example, the World Bank Global Financial Inclusion (Global Findex) database (Demirgüç-Kunt and Klapper, 2012) provides a wealth of user-side data on access, including on the extent of mobile banking in Africa. However, these data are only available for 2011 and 2014 and cannot be used for the sub-indices measuring access because they do not cover a sufficiently long time period.

Financial institutions efficiency sub-index relies on three aspects of bank efficiency: (i) efficiency in intermediating savings to investment, as measured by the net interest margin (the accounting value of bank’s net interest revenue as a share of its average interest-bearing assets) and lending-deposit spread; (ii) operational efficiency measures, such as non-interest income to total income and overhead costs to total assets; and (iii) profitability measures, such as return on assets and return on equity. As with the other dimensions, these are relatively crude measures of efficiency. For example, efficient financial institutions tend to be more profitable, but this relationship is not necessarily one for one, e.g. inefficient institutions can report profits when they operate in an economic upswing, while otherwise efficient institutions when hit by an adverse shock may generate losses.

We chose not to include in the efficiency sub-index indicators of microstructure, such as banking system concentration ratios – Herfindahl index or the share of top three banks in total banking system assets. They are important to assess the financial stability features as they provide a rough approximation for the potential impact in the case of a major financial disruption (Čihák and Schaeck, 2010). But there is no clear bottom line in the literature on whether more concentrated banking systems are more or less efficient. As surveyed in Berger et al. (2004), the findings for a range of efficiency indicators – loan pricing, interest margins, profitability, and firm access to credit, among others – are mixed and are not robust to controlling for institutional development, legal impediments to competition, and the different competitive effects of foreign-owned and state-owned banks.

Financial market indicators focus on stock market and debt market development. The depth sub-index includes the size of the stock market (capitalization, or the value of listed shares) and how active it is (stocks traded), the outstanding volume of international debt securities of sovereigns and international and domestic debt securities of financial and nonfinancial corporations. Corporate debt securities data are based on the nationality, rather than residence principle, to better align it with the sovereign debt data. We do not include the data on the outstanding volumes of domestic sovereign debt securities because these are provided to the BIS on a voluntary basis by the central banks and have low country coverage (18 countries at best). Dealogic corporate securities data have wider coverage than the BIS database and is therefore the preferred source of corporate debt data. It does not however allow a good distinction among the holders of corporate debt into domestic and external.

For the financial market access, we use the percentage of market capitalization outside of top 10 largest companies to proxy access to stock markets. A higher degree of stock market concentration should reflect greater difficulties in accessing the stock market for newer or smaller issuers. For bond market access, we use the number of financial and nonfinancial corporate issuers on the domestic and external debt market in a given year per 100,000 adults. This variable reflects the number of distinct issuers, such that repeat issuance by the same company in a given year is only counted once. It would be preferable to scale this variable by the total pool of potential issuers, but data limitations are a constraint. Dealogic only reports the number of companies that issue. Data on the number of listed domestic companies from the World Bank’s World Development Indicators only cover companies that issue on the domestic stock market and cover about 60 percent of our country-year sample. However, the correlation between this indicator and population size is 60 percent, which indicates that population size is a relatively good proxy.

Financial market efficiency sub-index relies on the stock market turnover ratio – the ratio of the value of stocks traded to stock market capitalization. A higher turnover should indicate higher liquidity and greater efficiency in the market. In the bond market, the most commonly used variable is the tightness of the bid-ask spread. Bloomberg data on the bid-ask spread in the sovereign bond market covers on average 37 countries (20 percent of the country sample) starting only in 2000. Given poor coverage, it is not used in the sub-index.

A number of variables are not included for conceptual reasons. The purpose of the index is to capture the key features of financial systems – how deep, accessible, and efficient they are. That is separate from capturing the underlying drivers of these features, such as the institutional, regulatory, and legal frameworks, or their outcomes, such as whether financial systems are more growth-enhancing or more stable. Therefore, the indices do not include potentially interesting indicators from the World Bank Doing Business database on the ease of getting credit (captured by the strength of legal rights, depth of credit information, and credit registry coverage), protection of minority investors, time and cost to enforce contracts, and the ease of resolving insolvency. Similarly, the database does not include financial stability indicators, such as z-scores, capital adequacy or liquidity ratios, and frequency of banking crises.

B. Treatment of missing data

There is a tradeoff between creating a comprehensive measure of financial development and data availability. The extent of missing data (Table 3) varies considerably across indicators. More data is available for a larger sample of countries in the most recent twenty years rather than earlier in the sample. Data coverage is strong for private credit, debt issuance, and financial institutions efficiency measures. It is weaker for non-bank financial institutions and other financial markets measures, especially in the LIDC sample. In some cases, such as financial institutions access measures, data are missing because they were not being collected before 2004 on a comprehensive basis. In other cases, lack of data indicates that markets may be missing. For example, only few of the LIDCs have developed their own domestic stock markets.

Table 3.Percent of Countries and Years with Data AvailabilityAverage by Decade and Income Group
Total number of countries/years183268968183*34
(percent of country-years)(percent of countries in the income group)(percent of country-years)
Financial Institutions Depth
Private sector credit to GDP7387939496838586
Pension fund assets to GDP0233423320215
Mutual fund assets to GDP2432394717216
Insurance premiums (life + non-life) to GDP060889167594354
Financial Institutions Access
Bank branches per 100,000 adults00559729272828
ATMs per 100,000 adults00479127252224
Financial Institutions Efficiency
Net interest margin054939566564855
Lending-deposits spread5378919390717876
Non-interest income to total income061939568585157
Overhead costs to total assets055949566564955
Return on assets056949566565055
Return on equity054939566564955
Financial Markets Depth
Stock market capitalization to GDP646616271501640
Stocks traded to GDP341596068461337
International debt securities of government to GDP02443465333425
Total debt securities of financial corporation to GDP02251585337728
Total debt securities of nonfinancial corporation to GDP027657754432036
Financial Markets Access
Percent of market capitalization outside of top 10 largest companies0422263013011
Total number of issuers of debt (domestic and external, fin. and non-fin. corporations)1927374068461329
Financial Markets Efficiency
Stock market turnover ratio (value traded/stock market capitalization)341596068461337
Source: IMF staff estimates.
Source: IMF staff estimates.

Several approaches are taken to address the missing data problem. Where data are not yet available for the latest year (e.g. 2013), the values are set equal to the latest available observations (e.g. 2012). This is the case, for example, with stock market capitalization and stocks traded data, which FinStats sources from the World Bank Development Indicators database and which are only available until 2012. If the data series is completely not available for a country, the entire series is set at zero, indicating this market does not exist or that its access or efficiency properties are very poor.

A more complicated case of missing variables arises when putting together series where database collection started at different points in time. For example, while observations on credit to GDP become available already in the 1960s, financial access data only started to be collected in 2004. This particular case of missing data can be treated in several ways: (i) treat the data as truly missing, excluding the series from the index average when the data are not available; (ii) treat the data as zero, assuming that the absence of data implies this market does not exist or its accessibility and efficiency are very poor; (iii) splice the two indices from before and after the data series becomes available.

As demonstrated in Figures 2 and 3, splicing is the preferred method. It avoids generating movements in the FD index that are unrelated to financial development, but are instead driven by the addition of new series. In Figure 2, series 2 is added to the index on which a country has worse performance relative to the other indicators and for which data are only available for the later part of the sample. The aggregate index should not drop as the series gets introduced (as it would if we were to treat the missing data as truly missing) because it is unlikely the country had a higher level of development before this market or data on this market appeared. In other words, the index should have started from a lower base (red or yellow line). In Figure 3, a missing series is added to the index on which a country has better performance relative to other indicators. Under missing or zero treatment, the index jumps as the series gets introduced, but it should not. This is a case where a country has a higher level of financial development on an indicator, but the data availability starts late. For example, just because the data on access to banking services is available starting in 2004, this does not mean that households did not have access to banking services before 2004.

Figure 2.Treatment of Missing Data: Example 1

One series is missing on which a country has worse performance

Source: IMF staff estimates.

Figure 3.Treatment of Missing Data: Example 2

One series is missing on which a country has better performance

Source: IMF staff estimates.

There is a simple intuitive explanation for what splicing accomplishes. If we were to assess in 2013, when all the data are available, the state of financial development across countries, we would do it by taking a weighted average across the performance on various indicators. We take this level of the index to determine the cross-country levels of financial development. When some data become unavailable as we go back in time, we move the index backwards using the average growth rate in the available series. In this way, we make an informed judgment as to whether data are missing but markets exist (for example, there was a bond issuance but there are no data on it), or whether missing data indicate non-existent markets (that is, there was no bond market). This method is preferred to some of the alternatives, such as fitting a trend line backwards into historical data, because it does not assume that financial development is a linear process. Indeed, like with economic developments, some countries go through stages of development, but then regress.

The only case where this approach could be inappropriate is when a big bang financial development event happens. Such a big bang event could be a country that develops a bond market from scratch with one or a few issuances in the first year such that: (i) their size is sufficiently large to place the country high in the cross-country rankings on the size of bond markets; (ii) the country rating is higher than its other financial development indicators and as a result there is an improvement in the composite index; and (iii) the country is able to sustain this size of the bond market going forward. Such a scenario would justify having a discrete jump in the index. In practice, such cases are rare in the data. In the last ten years, a typical first-time sovereign issuance averaged four percent of GDP (Guscina, Pedras, and Presciuttini, 2014), in line with the average in this dataset.

In the data set, splicing is applied at the level of the raw data. First, we identify series with data missing in the earlier years. Then missing data are filled in retrospectively, starting from the first available observation and applying the average growth rate of the other indicators with data available for previous years. This procedure is applied iteratively, using first the growth rates of similar series within the particular sub-index (e.g. financial institutions depth), then the growth rates of series for the same type of financial services provider (e.g. financial institutions), and only then across providers (from financial institutions to financial markets). The only series that are not used in this procedure are the two profit indicators. Given that they span the negative and positive range, the growth rates of these series would overstate movement in the other indicators.

It is very important to stress that the goal of this exercise is not to create artificial data. The new indicators should not be and are not used as standalone series. Instead, the series are adjusted such that the indices that are based on these indicators reflect financial system development, rather than data availability. In practice, about 27 percent of our sample is reconstructed through splicing, and 32 percent of our sample consists of “missing” markets (Figure 4).

Figure 4.Databank Composition: Raw, Reconstructed, and Missing Data Shares

Source: IMF staff estimates.

C. Normalization and treatment of outliers

Each series is winsorized to prevent extreme values from distorting the 0-1 indicators. For example, a particularly large negative ROE during a crisis will cause a bunching of the rest of the 0-1 ratings for ROE around 1. To avoid that, each indicator is winsorized, with the 5th and 95th percentiles set at the cutoff levels, so as not to lose data. Global distribution – across countries and time – is assessed to determine the cutoff levels.

Winsorized indicators are then normalized between 0 and 1, using the min-max procedures (equations 1 and 2) to facilitate aggregation over variables expressed in different measurement units:

where x is the underlying raw data and Ix is the transformed continuous 0-1 indicator.

The procedure normalizes indicators to have an identical range [0, 1] by subtracting the minimum value and dividing by the range of the indicator values. It relates country performance on an indicator to the global minimum and maximum across all countries and years. Thus, the highest (lowest) value of a given variable across time and countries is equal to one (zero) and all other values are measured relative to these maximum (minimum) values. For some series – net interest margin, lending-deposits spread, noninterest income to total income, and overhead costs to total assets – a higher value indicates a worse performance on efficiency. For these cases, the ratings are rescaled according to the min max formula 2 so that a higher value indicates greater financial development. The Human Development Index is one example of an index using the min-max normalization. See OECD (2008) for alternative normalization methods. The more common methods are standardization, the min-max, and the distance to a reference point.

D. Functional form of the aggregator

Indicators are then aggregated into the six sub-indices at the bottom of the pyramid in Figure 1. The aggregation is a weighted linear average of the underlying series, where the weights are obtained from principal component analysis, reflecting the contribution of each underlying series to the variation in the specific sub-index. All of the sub-indices are then re-normalized using equation 1, so that there range is between 0 and 1.

where FIj and FMj stand in turn for financial institutions depth (FID), access (FIA), efficiency (FIE), and for financial markets depth (FMD), access (FMA), efficiency (FME).

Sub-indices are aggregated into higher-level indices using the same procedure as above, culminating at the most aggregated level in the FD index. The FI, FM, and FD indices are again re-normalized, so that there range is between 0 and 1.

The linear functional form of the aggregator is best suited for the data with a significant share of zero or close to zero observations. Linear aggregation assumes full compensability, such that poor performance in some indicators can be compensated for by sufficiently high values in other indicators. In other words, it assumes that the indicators are perfect substitutes. An alternative aggregation method could be a geometric mean (equation 5), which allows for imperfect substitutability among indicators. Under geometric aggregation, higher financial efficiency, for example, does not fully compensate for low financial depth. As a result, a country with a more unequal distribution of indicator scores would receive a lower index rating (Figure 5). While an attractive concept, for our dataset, geometric averaging introduces a substantial zero bias in the indicator ratings (Figure 6). This is due to the fact that zero or close to zero indicator ratings drive the multiplicative averaging down to zero. This is not acceptable for conceptual reasons since the penalty for underperformance on one indicator appears to be too large (Luxembourg example below). In addition, by introducing a large number of close to zero observations in the final index geometric average reduces variability in the final sample, which limits its usefulness for research.

Figure 5.Linear Versus Geometric Aggregation: Hypothetical Example, Equal Weights

Source: IMF staff estimates.

Figure 6.Linear Versus Geometric Aggregation: 2013 FD Index Ratings

Source: IMF staff estimates.

A particularly stark example is Luxembourg’s FD rating (Figure 7). Luxembourg final FD score for 2013 would undergo the largest change if a geometric rather than linear aggregation were to be adopted. This is driven by financial market development (FM) rating, especially the one on financial markets depth (FMD). Luxembourg ranks the highest on the depth of its stock market and corporate debt market and is intermediate on government debt. But given that its stock market is relatively less traded, that particular indicator receives a very low normalization rating of 0.003. As a result, the FMD score for Luxembourg drops from 0.75 to 0.25 under geometric averaging and its ranking drops 29 places down. Given that other aspects of financial markets in Luxembourg are highly developed, it seems extreme to assign such a high weight to underperformance on one indicator out of five in assessing the depth of its financial markets.

Figure 7.Luxembourg Example: Index Ratings for 2013

Source: IMF staff estimates.

Note that the particular needs of geometric averaging require a different normalization of data and several other adjustments to make the results meaningful. For geometric averaging, the distance to a reference point, instead of the min-max procedure, is used for normalization (equation 6), because it is centered on 1 and does not give rise to zero indicator ratings. For indicators where an increase indicates a worsening performance (some of the banking efficiency indicators), the second functional form is used. Observations with zeros in raw data are replaced with the minima observed for that indicator. The scales of ROA and ROE are moved uniformly into the non-negative territory as geometric averaging does not allow negative values.

In addition to being the more appropriate method, linear aggregation is simpler to implement and interpret. In particular, the contribution of changes in each indicator to the changes in the FD index under linear aggregation is its weight. In other words, an additive aggregation function permits the assessment of the marginal contribution of each variable separately. In the case of a geometric mean, the contribution of changes in an indicator to changes in the index is more complex. It will depend on the level of other indicators, which may hinder the ease of interpretation.

E. Weights

When used in a benchmarking framework, weights can have a significant effect on the overall composite indicator and country rankings. A number of weighting techniques exist (see OECD, 2008 for an overview). Some are derived from statistical models, such as factor analysis, others from participatory methods, like analytical hierarchy process. Regardless of which method is used, weights are essentially value judgments. While some analysts might choose weights based only on statistical methods, others might reward components that are deemed more influential, depending on expert opinion, to better reflect policy priorities or theoretical factors.

For the FD index, this paper relies on a statistical method – the principal component analysis (PCA) – so as not to prejudge the importance of particular indicators in measuring financial development. Principal component analysis groups together individual indicators which are collinear to form a composite indicator that captures as much as possible of the information common to individual indicators. The idea is to account for the highest possible variation in the indicator set using the smallest possible number of factors. As a result, the composite index no longer depends upon the dimensionality of the data set but rather is based on the statistical dimensions of the data.

Sub-indices are constructed as weighted averages of the normalized series, where the weights are squared factor loadings (such that their sum adds up to 1) from principal component analysis of the underlying series. Factor loadings are coefficients that relate the observed variables to the principal components, or factors. The square of factor loadings represents the proportion of the total unit variance of the indicator which is explained by the factor. The series that contributes more to the direction of common variation in the data gets a higher weight. Weighting intervenes only to correct for overlapping information between two or more correlated indicators and is not a measure of the theoretical importance of the associated indicator.

The factor loadings on the first principal component are chosen as weights (Figure 8). Given the wide ranging nature of the exercise, the first principal component can be interpreted to summarize the latent information on the degree of financial development. Depending on the sub-index, it sums up the information on financial depth, access, and efficiency and embodies between 51 and 92 percent of the variance in the sub-index data (Table 4). The other principal components within the sub-index could reflect latent information on broader issues relevant for financial systems, such as governance and regulation or structural features.

Figure 8.Principal Component Analysis: Normalized Weights

Source: IMF staff estimates.

Table 4.Share of Variance Explained by PCA Components
Financial InstitutionsFinancial MarketsSub-indices
Source: IMF staff estimates.
Source: IMF staff estimates.

To summarize, PCA is done by pooling together all series in a particular sub-index across all countries (LIDC, EM, AM) and all years (1980-2013) to find the linear combination in the direction of the largest variation. A higher weight is given to a series that contributes more to the direction of common variation. Then sub-indices are combined into higher indices using the same procedure.

As Figure 8 shows, banking system credit to the private sector, while still a relevant component of financial development, has a weight of 0.25 within the depth subcomponent of FI, which in turn has a weight of less than 0.40 in the FI subcomponent. In other words, bank credit still plays an important role, reflecting the role of banks in many financial systems, but it is far from being the only driver of the results.

F. Putting it all together

To summarize, the procedure is as follows: (i) apply missing data treatment to actual data; (ii) winsorize to set the 5th and 95th percentiles at the cutoff levels to avoid extreme observations driving the best and worse scores; (iii) construct a relative ranking of countries for each indicator using the min-max procedure, where higher value indicates greater financial depth; (iv) construct sub-indices as weighted average of the underlying series, where the weights are squared factor loadings (sum to 1) from principal component analysis of the underlying series; (v) combine sub-indices into higher indices via a similar procedure.

The result is a relative ranking of countries on depth, access, and efficiency of financial institutions and financial markets, on the development of financial institutions and markets, and on the overall level of financial development. Figure 9 gives a world view of the state of financial development in 2013. Financial market development is low in Africa, and more advanced in Russia and China. See Annexes 1-3 for exact numbers behind these figures and the depth, access, and efficiency rankings that drive them and Table 5 for summary statistics.

Figure 9.World Map of Financial Development, 2013

Source: IMF staff estimates.

Table 5.Summary Statistics of the Financial Development Index
Var.ObsMeanMedianSt. Dev.MinMaxVar.ObsMeanMedianSt. Dev.MinMax
All countriesAdvanced Markets
Emerging MarketsLow-Income and Developing Countries
Source: IMF staff calculations.Note: FD = financial development; FI = financial institutions; FM = financial markets;FID = financial institutions depth; FIA = financial institutions access; FIE = financial institutions efficiency;FMD = financial markets; FMA = financial markets access; FME financial markets efficiency.
Source: IMF staff calculations.Note: FD = financial development; FI = financial institutions; FM = financial markets;FID = financial institutions depth; FIA = financial institutions access; FIE = financial institutions efficiency;FMD = financial markets; FMA = financial markets access; FME financial markets efficiency.

The indices are an improvement over the traditional measures of financial development. Conceptually, they incorporate information on a broader range of financial development features for a wider array of financial agents. Indeed, as Figures 10 and 11 show, while the indices are correlated with the traditional measures – private credit to GDP and stock market capitalization to GDP – the correlation is not one for one, e.g. the indices contain more information.

Figure 10.Correlation of FD Index with Traditional Measures: Private Credit to GDP

Source: IMF staff estimates.

Figure 11.Correlation of FD Index with Traditional Measures: Stock Market Capitalization/GDP

Source: IMF staff estimates.

A few specific country examples could help interpret the FD rankings. Among advanced markets, it may seem surprising that the global financial centers, such as the United Kingdom and Hong Kong, rank somewhat lower than Korea and Australia on financial market development in 2013 (Annex 1). While the United Kingdom has the deepest financial markets among these four countries, it ranks the lowest in this group of four on financial market access and efficiency. In the UK, market capitalization outside of top 10 companies in 2013 is 30 percent, compared to 50 percent in Australia and 38 percent in Korea. Corporate issuance per 100,000 adults stands at 0.6 versus 0.9 in Australia and 1 in Korea. Finally, its stock market turnover is 84 percent, while it is 85 percent in Australia and 139 percent in Korea. Similarly, while Hong Kong ranks highly on financial market efficiency, its overall FM indicator is brought down by lower depth and access.

A similar picture holds in other regions and income groups. Trinidad and Tobago, the wealthiest and most developed nation in the Caribbean region, receives a lower FD rating compared to St. Kitts and Nevis. This is due to lower ratings on financial institutions development. While financial institutions are larger in Trinidad and Tobago, they rank lower on access and efficiency measures. In terms of branches and ATMs per 100,000 adults, Trinidad and Tobago has 13 and 41, while St. Kitts and Nevis has 55 and 107. In terms of efficiency, Trinidad and Tobago has higher net interest margins and overhead costs at five and four percent, compared to St. Kitts and Nevis’ 0.7 and 1.3 percent.

These examples help highlight the fact that financial system development needs to be assessed in a comprehensive way. Countries that we would typically associate with the most developed status either globally or regionally due to the size of their financial institutions and markets may not necessarily be so, at least up to a margin, once we take into account how accessible their financial systems are to households and corporates and how efficient they are in delivering their services.

III. Landscape of Financial Development

The evolution of the FD index over the sample period (1980–2013) shows a pattern that generally confirms priors (Figure 12). Overall, financial development has progressed quite noticeably in both advanced economies (AEs) and emerging markets (EMs), and to a lesser extent in low-income developing countries (LIDCs). However, as one would expect, the gap between the first two groups widened significantly between the mid-1990s and early 2000s, reflecting particularly rapid growth in AEs’ financial systems. This episode marks the “Greenspan Era” in the United States, a period when European cross-border banking expanded considerably, as did investment banking and internet banking.2 On the other hand, during this period financial development proceeded more moderately in EMs and was relatively stagnant in LIDCs. The gap in financial development between the AEs and EMs has subsequently declined after the global financial crisis, reflecting deleveraging in AEs.

Figure 12.Financial Development through Time

Source: IMF staff estimates

A snapshot comparison across peer groups presents quite a diverse picture (Figure 13). In particular, the “gap” in financial development between AEs and EMs differs across the various dimensions of financial development highlighted in the figure. For example, EMs are closer to AEs in financial markets development rather than in financial institutions. Also, despite lower depth, the efficiency of EM and LIDC financial institutions is relatively high. Finally, access seems to be particularly low in LIDCs, making this an area of potential improvement.

Figure 13.Financial Development Index: Peer Group Averages

Source: IMF staff estimates

Looking at individual country rankings as of 2013, there is substantial variation in financial development within and across income groups (Figure 14). Some large EMs, such as Malaysia, Brazil, and South Africa, have higher levels of financial development than certain AEs, such as New Zealand and Greece. Also, several EMs, such as Tunisia and Armenia, have lower levels of financial development than some LIDCs, such as Mongolia and Bangladesh.

Figure 14.Financial Development Index: Selected Countries, 2013

Source: IMF staff estimates

IV. Caveats and Limitations

A challenge for all empirical literature is that the broad measures of financial development capture only partially the various functions of finance, such as its ability to facilitate risk management, exert corporate control, pool savings, allocate capital to productive investment, and facilitate exchange of goods (Levine 2005). This paper addresses the challenge by relying on a broad set of indicators to develop a more comprehensive index of financial development.

There are a number of limitations to the FD index that need to be taken into account when interpreting empirical results. On the data coverage side, it was not possible to find sufficiently extensive country and time period data on some institutions and activities. One example is shadow banks, whose importance has been rising in a number of EMs, with associated risks (for a recent analysis on this topic for a smaller country sample, see IMF 2014). Different forms of financial payments, such as credit transfers, direct debits, and mobile banking, are undeniably relevant aspects of depth and access in many countries, but indicators of these are currently not available on a sufficiently long time horizon to warrant inclusion in the FD index. Other potentially relevant features of financial development – such as the diversity in the types of financial intermediaries and the organizational complexity of institutions and instruments – are not incorporated in the index.

There are also caveats on the conceptual side. First, by design the FD index only captures the characteristics of financial systems (depth, access, efficiency). It does not include their underlying drivers (such as the institutional, regulatory, and legal frameworks) or outcomes (financial stability measures). Second, some of the measures that the index uses may overstate the true level of financial development. In some countries some of the efficiency measures could reflect government controls, for example on lending and deposit rates, which may inflate efficiency ratings. Finally, researchers need to benchmark the FD index vis-à-vis its determinants, e.g. various country characteristics (see for more detailed guidance Feyen, Kibuuka, and Sourrouille, 2014). Higher FD ranking may not necessarily be a good thing, but may instead indicate that a country’s financial system is stretched beyond it structural and regulatory capabilities, with negative implications for growth and stability.

While there are challenges in constructing the index, it serves as an important step toward measuring financial development more comprehensively than before and should aid researchers studying the various relationships between financial development and economic outcomes. We strive to continue improving the financial development indices as new information becomes available.


    Aizenman, Joshua, YothinJinjarak, and DonghyunPark,2015. “Financial Development and Output Growth in Developing Asia and Latin America: A Comparative Sectoral Analysis.” NBER Working Paper 20917 National Bureau of Economic Research, Cambridge, Massachusetts.

    • Search Google Scholar
    • Export Citation

    Amidžić, Goran, AlexanderMassara, and AndréMialou,2014. “Assessing Countries’ Financial Inclusion Standing – A New Composite Index.” IMF Working Paper 14/36. Washington: International Monetary Fund (February).

    • Search Google Scholar
    • Export Citation

    Arcand, Jean-Louis, EnricoBerkes, and UgoPanizza,2012. “Too Much Finance?” IMF Working Paper 12/161. Washington: International Monetary Fund (June).

    • Search Google Scholar
    • Export Citation

    Berger, Allen, AsliDemirgüç-Kunt, RossLevine, and JosephHaubrich,2004. “Bank concentration and competition: An evolution in the making.” Journal of Money, Credit and Banking. 36: 433–451

    • Crossref
    • Search Google Scholar
    • Export Citation

    Bernanke, Ben, MarkGertler, and SimonGilchrist,1999. “The Financial Accelerator in a Quantitative Business Cycle Framework,” NBER Working Paper 6455, National Bureau of Economic Research, Cambridge, Massachusetts.

    • Search Google Scholar
    • Export Citation

    BIS debt securities statistics, 2015. Bank for International Settlement.

    Camara, Noelia and DavidTuesta,2014. “Measuring Financial Inclusion: A Multidimensional Index,” Working Papers 1426, BBVA Bank, Economic Research Department.

    • Search Google Scholar
    • Export Citation

    Cardarelli, Roberto, SelimElekdag, and SubirLall,2008. “Financial Stress and Economic Downturns,” World Economic Outlook, October 2008 Issue, International Monetary Fund, Chapter 4, 129–158.

    • Search Google Scholar
    • Export Citation

    Cardarelli, Roberto, SelimElekdag, and SubirLall,2009. “Financial Stress, Downturns, and Recoveries,” IMF Working Papers 09/100. Washington: International Monetary Fund.

    • Search Google Scholar
    • Export Citation

    Čihák, Martin, AsliDemirgüç-Kunt, ErikFeyen, and RossLevine,2012. “Benchmarking Financial Development Around the World.” World Bank Policy Research Working Paper 6175. World Bank, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Čihák, Martin, AsliDemirgüç-Kunt, ErikFeyen, and KlausSchaeck.2010. “How Well do Aggregate Prudential Ratios Identify Banking System Problems?” Journal of Financial Stability, vol. 6(3), pp. 130–144

    • Crossref
    • Search Google Scholar
    • Export Citation

    Dabla-Norris, Era, and NarapongSrivisal,2013. “Revisiting the Link between Finance and Macroeconomic Volatility.” IMF Working Paper 13/29. Washington: International Monetary Fund (January).

    • Search Google Scholar
    • Export Citation

    Dealogic corporate debt database.

    Demirgüç-Kunt, Asli, and RossLevine,2009. “Finance and Inequality: Theory and Evidence,” Annual Review of Financial Economics, Annual Reviews, vol. 1(1), pages 287–318, November.

    • Crossref
    • Search Google Scholar
    • Export Citation

    Demirgüç-Kunt, Asli, and LeoraKlapper.2012. “Measuring Financial Inclusion: The Global Findex,” Policy Research Working Paper 6025, World Bank, Washington, DC.

    • Search Google Scholar
    • Export Citation

    Feyen, Erik, KatieKibuuka, and DiegoSourrouille,2014. “FinStats 2015: A ready-to-use tool to benchmark financial sectors across countries and time”, World Bank mimeo.

    • Search Google Scholar
    • Export Citation

    Financial Access Survey, 2014. International Monetary Fund.

    FinStats, 2015. World Bank.

    Global Findex Database, 2014. World Bank.

    Guscina, Anastasia, GuilhermePedras, and GabrielPresciuttini,2014. “First-Time International Bond Issuance—New Opportunities and Emerging Risks” IMF Working Paper 14/127. Washington: International Monetary Fund (July).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF), 2014. “Shadow Banking Around the Globe: How Large and How Risky?” In Global Financial Stability Report, Washington, October 2014.

    • Search Google Scholar
    • Export Citation

    Levine, Ross,2005. “Finance and Growth: Theory and Evidence,” Handbook of Economic Growth, in: PhilippeAghion & StevenDurlauf (ed.), Handbook of Economic Growth, edition 1, volume 1, chapter 12, pages 865–934 Elsevier.

    • Crossref
    • Search Google Scholar
    • Export Citation

    OECD/JRC, 2008, Handbook on constructing composite indicators. Methodology and user guide. OECD Publisher, Paris.

    Rajan, Raghuram G. and LuigiZingales,1998. “Financial Dependence and Growth”, American Economic Review, 88: 559–586.

    Sahay, Ratna, Martin Cihák, Papa N’Diaye, Adolfo Barajas, Ran Bi, Diana Ayala, Yuan Gao, Annette Kyobe, Lam Nguyen, ChristianSaborowski, KatsiarynaSvirydzenka, and SeyedReza Yousefi,2015. “Rethinking Financial Deepening: Stability and Growth in Emerging Markets.” IMF Staff Discussion Note 15/08. Washington: International Monetary Fund (May).

    • Search Google Scholar
    • Export Citation

    UNDP, 2014, Human Development Report, New York: Palgrave Macmillan.

Annex 1. 2013 Country Rankings on Financial Development
Financial Development IndexFinancial Insitutions IndexFinancial Markets Index
1Switzerland0.9511Switzerland1.0001United States0.903
2Australia0.8902Luxembourg0.8932Korea, Republic of0.902
3United Kingdom0.8823France0.8923Switzerland0.883
4United States0.8774United Kingdom0.8924Australia0.873
5Spain0.8605Canada0.8905Hong Kong0.869
6Korea, Republic of0.8546Japan0.8906United Kingdom0.855
9Hong Kong0.8279Denmark0.8569Norway0.764
13Sweden0.74913United States0.83313Germany0.731
16Singapore0.73116Korea, Republic of0.78916Singapore0.695
18Netherlands0.71018Malta0.77818Saudi Arabia0.653
19Austria0.70719Hong Kong0.76919Russian Federation0.623
20Belgium0.69320Israel0.76520China, Mainland0.622
22Norway0.67922New Zealand0.75122Malaysia0.617
26Thailand0.64526Bahamas, The0.72526Denmark0.602
28South Africa0.61828South Africa0.71328Greece0.540
29New Zealand0.60929Cyprus0.69929Belgium0.525
30Israel0.59630Netherlands0.69030South Africa0.511
32Russian Federation0.59232Bulgaria0.68232U.A.E.0.488
33China, Mainland0.57233Thailand0.66633Portugal0.473
35Cyprus0.55635Slovenia0.65335New Zealand0.456
36Chile0.54536St. Kitts and Nevis0.64336Qatar0.450
38Saudi Arabia0.53038Antigua & Barbuda0.61838India0.431
40Bahamas, The0.47540Finland0.59740Israel0.415
46Barbados0.43546Russian Federation0.54946Colombia0.333
Financial Development IndexFinancial Insitutions IndexFinancial Markets Index
47Jordan0.41447Slovak Republic0.54747Barbados0.328
51India0.39251St. Lucia0.53651Egypt0.281
53Mauritius0.38953Czech Republic0.53353Slovenia0.267
55St. Kitts and Nevis0.36655Morocco0.52855Moldova0.250
57Czech Republic0.36057China, Mainland0.51157Morocco0.243
59Brunei Darussalam0.33659Costa Rica0.50359Bahamas, The0.216
62Trinidad & Tobago0.32862Lithuania0.49162Jamaica0.187
63Indonesia0.32263Ecuador0.48963Sri Lanka0.185
64Lebanon0.32164Namibia0.48864Iran, I. Rep. Of0.182
65Argentina0.31465Trinidad & Tobago0.48865Brunei Darussalam0.181
66Slovak Republic0.31466Brunei Darussalam0.48566Czech Republic0.181
68Antigua & Barbuda0.31268Cape Verde0.48068Trinidad & Tobago0.161
69Kazakhstan0.31169Turkey0.47469Papua New Guinea0.155
70Bahrain0.30470Macedonia, FYR0.46870Cote D’Ivoire0.138
71Latvia0.29871Bosnia and Herzegovina0.46471Panama0.138
75St. Lucia0.28875Mexico0.44375Mongolia0.105
76Costa Rica0.28476Guatemala0.44376Vietnam0.103
80Sri Lanka0.27080Georgia0.42680Liberia0.088
82Ecuador0.25882El Salvador0.41782Burundi0.085
83Ukraine0.25783Armenia0.41683St. Kitts and Nevis0.081
85Venezuela0.25585St. Vincent and the Gren0.40285Uzbekistan0.079
86Macedonia, FYR0.25186Uruguay0.40286Venezuela0.079
87Iran, I. Rep. Of0.24987Tunisia0.40087Slovak Republic0.074
89El Salvador0.24789Saudi Arabia0.39689El Salvador0.071
91Cape Verde0.24391Suriname0.39091Bulgaria0.071
92Georgia0.23992Macao SAR, China0.38892Honduras0.065
Financial Development IndexFinancial Insitutions IndexFinancial Markets Index
93Tunisia0.23993Bolivia0.38793Costa Rica0.061
96Bosnia and Herzegovina0.23696Vietnam0.36496Lithuania0.051
98Belize0.22398Sri Lanka0.34998Georgia0.048
106St. Vincent and the Gren0.203106Moldova0.338106Zambia0.035
107Albania0.200107Serbia0.334107St. Lucia0.034
110Suriname0.197110Iran, I. Rep. Of0.312110Macedonia, FYR0.030
111Macao SAR, China0.196111Aruba0.312111Bhutan0.029
114Papua New Guinea0.187114Kenya0.299114Turkmenistan0.025
117Nepal0.176117Dominican Republic0.293117Malawi0.022
119Cote D’Ivoire0.168119Belarus0.289119Kyrgyz Republic0.021
120Azerbaijan0.160120Azerbaijan0.287120Dominican Republic0.020
122Dominican Republic0.158122Jamaica0.283122Yemen0.018
133Lesotho0.136133Sao Tome and Principe0.238133Madagascar0.009
134Gabon0.133134Micronesia, Fed. Sts.0.234134Swaziland0.006
137Algeria0.128137Nigeria0.225137Sierra Leone0.006
138Zambia0.128138Kyrgyz Republic0.225138Nicaragua0.006
Financial Development IndexFinancial Insitutions IndexFinancial Markets Index
139Kyrgyz Republic0.124139Laos0.224139Burkina Faso0.005
141Sao Tome and Principe0.120141Senegal0.218141Cameroon0.003
142Micronesia, Fed. Sts.0.118142Mozambique0.218142Togo0.003
143Ghana0.118143Burkina Faso0.217143Bosnia and Herzegovina0.003
144Burundi0.117144Papua New Guinea0.215144Algeria0.002
148Burkina Faso0.112148Myanmar0.203148Cape Verde0.002
153Tanzania0.103153Cameroon0.195153Congo, Dem. Rep. of0.000
154Mauritania0.102154Cote D’Ivoire0.194154French Polynesia0.000
155Benin0.102155Solomon Islands0.193155South Sudan0.000
157Cameroon0.100157Tanzania0.187157Timor Leste0.000
159Solomon Islands0.098159Gambia, The0.169159Equatorial Guinea0.000
160Uganda0.096160Haiti0.168160Marshall Islands0.000
162Malawi0.093162Turkmenistan0.164162Congo, Republic of0.000
165Gambia, The0.085165Congo, Dem. Rep. of0.159165Gambia, The0.000
166Haiti0.085166Rwanda0.156166Solomon Islands0.000
169Congo, Dem. Rep. of0.080169Uganda0.154169Myanmar0.000
171Madagascar0.079171Burundi0.146171Micronesia, Fed. Sts.0.000
172Eritrea0.079172Guinea0.145172Sao Tome and Principe0.000
174Sierra Leone0.063174Sierra Leone0.120174Aruba0.000
175Congo, Republic of0.059175Congo, Republic of0.116175Maldives0.000
177Marshall Islands0.050177Marshall Islands0.099177Macao SAR, China0.000
178Equatorial Guinea0.041178Equatorial Guinea0.081178Suriname0.000
179Comoros0.035179Comoros0.068179St. Vincent and the Gren0.000
180Timor Leste0.024180Timor Leste0.048180Vanuatu0.000
182South Sudan0.012182South Sudan0.024182Grenada0.000
183French Polynesia0.000183French Polynesia0.000183Antigua & Barbuda0.000
Source: IMF staff estimates.
Source: IMF staff estimates.
Annex 2. 2013 Country Rankings on Financial Institutions Depth, Access, Efficiency
Financial Institutions DepthFinancial Institutions AccessFinancial Institutions Efficiency
1Ireland1.0001St. Kitts and Nevis1.0001Greece0.784
2Denmark1.0002Brazil1.0002New Zealand0.751
3United Kingdom1.0003Portugal1.0003Japan0.749
4Switzerland1.0004Spain1.0004China, Mainland0.747
5Hong Kong0.9795Luxembourg0.9945Australia0.735
9South Africa0.8909France0.9229Estonia0.717
10United States0.81710Russian Federation0.91910Bahrain0.712
11Australia0.81311Belgium0.91311Korea, Republic of0.711
12Sweden0.80212Bahamas, The0.87812Malta0.710
14Japan0.77314United States0.87014Kuwait0.699
17Korea, Republic of0.72417Iceland0.83617Oman0.693
20Iceland0.68020Antigua & Barbuda0.80620Spain0.690
22Israel0.67822Mongolia0.74822Czech Republic0.690
23Belgium0.67423United Kingdom0.74323U.A.E.0.690
24Finland0.65824Malta0.71824Macao SAR, China0.689
28New Zealand0.61228Korea, Republic of0.70028Vietnam0.684
29Portugal0.60429Colombia0.69529Slovak Republic0.681
30Brazil0.58530New Zealand0.69230Nepal0.681
35St. Lucia0.47035Brunei Darussalam0.65335Denmark0.674
37Trinidad & Tobago0.46137Poland0.63737France0.671
38Bahamas, The0.43238Thailand0.63338Canada0.668
40China, Mainland0.41340Denmark0.60840Sri Lanka0.667
41Jordan0.39941Costa Rica0.59341Jordan0.664
44Namibia0.37044Turkey0.57844Bahamas, The0.654
45Greece0.36645Cape Verde0.57345Mauritius0.650
46Hungary0.36546Slovak Republic0.56846Chile0.649
Financial Institutions DepthFinancial Institutions AccessFinancial Institutions Efficiency
48Antigua & Barbuda0.35048Iran, I. Rep. Of0.55648Myanmar0.645
50El Salvador0.32050Lithuania0.54850Poland0.644
51Czech Republic0.31751Bosnia and Herzegovina0.54751Venezuela0.644
55Slovak Republic0.29555Macedonia, FYR0.52655Suriname0.634
58Panama0.28558Czech Republic0.50458Switzerland0.625
60Colombia0.27660Hong Kong0.49560Ethiopia0.623
63Costa Rica0.26963Uzbekistan0.47563Brunei Darussalam0.617
65Mexico0.26465St. Lucia0.46065El Salvador0.612
68Peru0.24468Mauritius0.45068South Africa0.609
70St. Vincent and the Gren0.23870Hungary0.44670St. Kitts and Nevis0.605
71Tunisia0.22771Saudi Arabia0.44071Macedonia, FYR0.604
72Bahrain0.22172Argentina0.42872United Kingdom0.602
76Bosnia and Herzegovina0.20876South Africa0.41676Burkina Faso0.596
77Argentina0.20677Macao SAR, China0.41577Cape Verde0.596
79Macedonia, FYR0.19879Albania0.40179Djibouti0.591
84Cape Verde0.18684Barbados0.39184Armenia0.578
88Philippines0.18388St. Vincent and the Gren0.38588Turkmenistan0.572
89Russian Federation0.17889Jordan0.37989Saudi Arabia0.564
91Serbia0.16791Uruguay0.36091Trinidad & Tobago0.562
92Mongolia0.16592Sao Tome and Principe0.35692Dominica0.561
Financial Institutions DepthFinancial Institutions AccessFinancial Institutions Efficiency
94Paraguay0.16494Suriname0.34894Hong Kong0.560
95Indonesia0.16095Trinidad & Tobago0.33795Colombia0.552
98St. Kitts and Nevis0.14998Indonesia0.32198Fiji0.544
100Togo0.143100China, Mainland0.316100Bosnia and Herzegovina0.541
101Aruba0.142101Finland0.313101St. Lucia0.538
102Saudi Arabia0.137102Maldives0.308102Romania0.536
104Ecuador0.135104Dominican Republic0.289104Ireland0.530
106Malawi0.133106Sri Lanka0.284106Hungary0.529
107Dominican Republic0.133107Bhutan0.278107Swaziland0.528
108Oman0.131108Azerbaijan0.276108Papua New Guinea0.528
109Moldova0.131109El Salvador0.273109Laos0.523
110Armenia0.123110Belarus0.266110Congo, Dem. Rep. of0.523
111Djibouti0.123111Botswana0.246111Costa Rica0.522
113Albania0.118113Marshall Islands0.241113St. Vincent and the Gren0.516
114Mozambique0.112114Micronesia, Fed. Sts.0.238114Turkey0.515
120Sri Lanka0.109120India0.198120Angola0.508
122Guatemala0.107122Kyrgyz Republic0.190122Lesotho0.506
123Brunei Darussalam0.105123Gabon0.188123Senegal0.504
125Papua New Guinea0.100125Nicaragua0.156125Kenya0.502
127Maldives0.095127Libya0.147127Antigua & Barbuda0.500
129Cote D’Ivoire0.093129Nepal0.135129Peru0.496
131Nigeria0.086131Solomon Islands0.133131Bolivia0.494
133Iran, I. Rep. Of0.082133Tajikistan0.125133Paraguay0.488
134Burkina Faso0.081134Equatorial Guinea0.121134Eritrea0.486
135Benin0.081135Bangladesh0.120135Micronesia, Fed. Sts.0.485
Financial Institutions DepthFinancial Institutions AccessFinancial Institutions Efficiency
140Sao Tome and Principe0.069140Zambia0.102140Samoa0.482
141Tonga0.068141Gambia, The0.096141Nigeria0.481
145Syria0.065145Algeria0.088145United States0.479
146Pakistan0.064146Cote D’Ivoire0.082146Sudan0.473
148Algeria0.060148Mali0.082148Kyrgyz Republic0.461
150Congo, Republic of0.059150Syria0.078150Ghana0.460
151Macao SAR, China0.057151Togo0.076151Niger0.459
153Haiti0.056153Timor Leste0.071153Guinea0.448
154Gambia, The0.055154Papua New Guinea0.062154Tonga0.436
155Rwanda0.054155Malawi0.059155Cote D’Ivoire0.434
156Solomon Islands0.052156Congo, Republic of0.058156Togo0.430
158Niger0.051158Tanzania0.056158Dominican Republic0.422
159Comoros0.051159Uganda0.055159Solomon Islands0.416
164Kyrgyz Republic0.042164Guinea-Bissau0.048164Gambia, The0.381
166Libya0.037166Burundi0.041166Russian Federation0.345
167Guinea-Bissau0.031167Burkina Faso0.040167Rwanda0.344
168Equatorial Guinea0.026168Cameroon0.037168Sierra Leone0.344
170Sierra Leone0.023170Sierra Leone0.034170Tajikistan0.335
173Yemen0.021173Madagascar0.028173Congo, Republic of0.242
174Congo, Dem. Rep. of0.019174Guinea0.026174Sao Tome and Principe0.222
175Liberia0.012175Niger0.024175Iran, I. Rep. Of0.175
176Uzbekistan0.011176South Sudan0.019176Comoros0.093
178South Sudan0.003178Congo, Dem. Rep. of0.011178Equatorial Guinea0.074
179French Polynesia0.000179Chad0.011179Timor Leste0.067
180Timor Leste0.000180French Polynesia0.000180South Sudan0.055
181Marshall Islands0.000181Eritrea0.000181French Polynesia0.000
183Micronesia, Fed. Sts.0.000183Bahrain0.000183Marshall Islands0.000
Source: IMF staff estimates.
Source: IMF staff estimates.
Annex 3. 2013 Country Rankings on Financial Markets Depth, Access, Efficiency
Financial Markets DepthFinancial Markets AccessFinancial Markets Efficiency
1Sweden0.9961Norway1.0001Saudi Arabia1.000
3United Kingdom0.9733Luxembourg1.0003China, Mainland1.000
4United States0.9714Switzerland0.9774Italy1.000
5Switzerland0.9705Austria0.9085Hong Kong1.000
6Spain0.9086Australia0.8356Korea, Republic of1.000
8Netherlands0.9028U.A.E.0.7648United States1.000
9Singapore0.8959Korea, Republic of0.7549Japan0.950
10Korea, Republic of0.89010Hong Kong0.73710Germany0.874
11Finland0.82011United Kingdom0.70811Russian Federation0.834
13Hong Kong0.81513Canada0.68713United Kingdom0.800
17Denmark0.73817United States0.66517Netherlands0.674
18South Africa0.73518Germany0.61118Thailand0.671
20Thailand0.70020New Zealand0.59220France0.632
23Ireland0.63523Russian Federation0.57623Bangladesh0.582
25Philippines0.62625Slovenia0.56025South Africa0.523
27China, Mainland0.59127Peru0.55127Hungary0.520
31Saudi Arabia0.51631Brunei Darussalam0.50031Singapore0.412
37Russian Federation0.44537Japan0.48737New Zealand0.314
39New Zealand0.42339Jordan0.47039Malaysia0.272
41Brazil0.40841Brazil0.43341Czech Republic0.257
42Papua New Guinea0.40642Israel0.42742U.A.E.0.241
44Jamaica0.38844Saudi Arabia0.42744Indonesia0.222
46Cote D’Ivoire0.34246Thailand0.40946Iran, I. Rep. Of0.169
Financial Markets DepthFinancial Markets AccessFinancial Markets Efficiency
51Bahamas, The0.28851Egypt0.33951Vietnam0.125
52Oman0.28252Sri Lanka0.33652Qatar0.116
54Trinidad & Tobago0.26054Bahamas, The0.32254Romania0.109
58Liberia0.23458Iran, I. Rep. Of0.26658Sri Lanka0.087
59Laos0.23459China, Mainland0.25059Estonia0.087
62Lebanon0.21362South Africa0.20762Guatemala0.061
63Burundi0.21263Trinidad & Tobago0.18163Slovenia0.059
64St. Kitts and Nevis0.20864Latvia0.17264Morocco0.059
65Czech Republic0.20065Estonia0.09865Uzbekistan0.056
69Malta0.17869Costa Rica0.07569Macedonia, FYR0.052
71Vietnam0.16571Czech Republic0.07171Bulgaria0.046
74El Salvador0.16274Slovak Republic0.06674Lebanon0.038
77Ukraine0.15377Lithuania0.03977Slovak Republic0.034
78Egypt0.14578Uzbekistan0.03778Kyrgyz Republic0.032
81Sri Lanka0.12281Croatia0.03081Latvia0.028
85Slovak Republic0.10985Dominican Republic0.02685Cote D’Ivoire0.022
86Iran, I. Rep. Of0.10586Azerbaijan0.02586Croatia0.022
87Georgia0.10187El Salvador0.02587Costa Rica0.019
88Uganda0.09488Kyrgyz Republic0.02488Bahrain0.018
90St. Lucia0.09090Bolivia0.02190Ghana0.016
Financial Markets DepthFinancial Markets AccessFinancial Markets Efficiency
93Costa Rica0.07993Ecuador0.01393Malawi0.014
97Lithuania0.06797Venezuela0.00997St. Kitts and Nevis0.008
98Uruguay0.06698Cote D’Ivoire0.00898Uruguay0.008
99Turkmenistan0.06699Mozambique0.00699Trinidad & Tobago0.007
101Latvia0.063101Romania0.006101El Salvador0.006
102Bolivia0.062102Algeria0.005102Papua New Guinea0.006
109Guyana0.054109French Polynesia0.000109Uganda0.001
110Zambia0.047110South Sudan0.000110Swaziland0.000
111Yemen0.047111Guinea-Bissau0.000111French Polynesia0.000
112Niger0.046112Timor Leste0.000112South Sudan0.000
114Gabon0.043114Equatorial Guinea0.000114Timor Leste0.000
115Brunei Darussalam0.040115Marshall Islands0.000115Comoros0.000
116Namibia0.039116C.A.R.0.000116Equatorial Guinea0.000
117Ethiopia0.038117Congo, Republic of0.000117Marshall Islands0.000
118Macedonia, FYR0.035118Sierra Leone0.000118C.A.R.0.000
119Dominican Republic0.031119Guinea0.000119Congo, Republic of0.000
120Mauritania0.030120Eritrea0.000120Sierra Leone0.000
123Ecuador0.029123Congo, Dem. Rep. of0.000123Madagascar0.000
125Guatemala0.027125Tajikistan0.000125Congo, Dem. Rep. of0.000
127Chad0.026127Gambia, The0.000127Tajikistan0.000
129Madagascar0.025129Niger0.000129Gambia, The0.000
133Senegal0.016133Solomon Islands0.000133Solomon Islands0.000
135Sierra Leone0.015135Cameroon0.000135Cameroon0.000
137Burkina Faso0.013137Benin0.000137Benin0.000
Financial Markets DepthFinancial Markets AccessFinancial Markets Efficiency
142Bosnia and Herzegovina0.007142Yemen0.000142Yemen0.000
143Kyrgyz Republic0.007143Burkina Faso0.000143Burkina Faso0.000
147Cape Verde0.005147Micronesia, Fed. Sts.0.000147Burundi0.000
148Algeria0.001148Sao Tome and Principe0.000148Micronesia, Fed. Sts.0.000
149Sudan0.001149Cambodia0.000149Sao Tome and Principe0.000
153Congo, Dem. Rep. of0.000153Lesotho0.000153Nicaragua0.000
154French Polynesia0.000154Libya0.000154Gabon0.000
155South Sudan0.000155Swaziland0.000155Lesotho0.000
157Timor Leste0.000157Tonga0.000157Djibouti0.000
159Equatorial Guinea0.000159Aruba0.000159Belarus0.000
160Marshall Islands0.000160Laos0.000160Tonga0.000
162Congo, Republic of0.000162Nepal0.000162Laos0.000
163Eritrea0.000163Samoa0.000163Dominican Republic0.000
164Haiti0.000164Papua New Guinea0.000164Maldives0.000
165Gambia, The0.000165Bhutan0.000165Azerbaijan0.000
166Solomon Islands0.000166Serbia0.000166Samoa0.000
167Kiribati0.000167Macao SAR, China0.000167Bhutan0.000
168Benin0.000168Suriname0.000168Macao SAR, China0.000
170Syria0.000170St. Vincent and the Gren0.000170Albania0.000
171Micronesia, Fed. Sts.0.000171Vanuatu0.000171St. Vincent and the Gren0.000
172Sao Tome and Principe0.000172Fiji0.000172Vanuatu0.000
174Aruba0.000174Bosnia and Herzegovina0.000174Belize0.000
175Maldives0.000175Cape Verde0.000175Bosnia and Herzegovina0.000
176Samoa0.000176Dominica0.000176Cape Verde0.000
177Macao SAR, China0.000177Macedonia, FYR0.000177Dominica0.000
179St. Vincent and the Gren0.000179St. Lucia0.000179St. Lucia0.000
181Dominica0.000181Moldova0.000181Antigua & Barbuda0.000
182Grenada0.000182Antigua & Barbuda0.000182Brunei Darussalam0.000
183Antigua & Barbuda0.000183St. Kitts and Nevis0.000183Bahamas, The0.000
Source: IMF staff estimates.
Source: IMF staff estimates.

This index was originally prepared for the IMF Staff Discussion Note “Rethinking Financial Deepening: Stability and Growth in Emerging Markets” (Sahay et al., 2015). I would like to thank Ross Levine, Ratna Sahay, Martin Čihák, Papa N’Diaye, Adolfo Barajas, and seminar participants at the IMF and the 2015 NBER Summer Institute’s International Finance & Macroeconomics Data Session for their useful comments and suggestions. Lam Nguyen, Diana Ayala, and Yuan Gao provided invaluable research assistance.


Figure 11 shows simple averages across countries, so the weight of the United States is relatively small. Also, direct cross-border bank lending is not captured by the index to the extent that it is not reflected in domestic credit provision.

Other Resources Citing This Publication