FACULTY OF ECONOMICS AND BUSINESS
The Impact of International Capital Flows on
Financial Stability
IE&B Master Thesis
Yanhong Lin (S2883627)
Y.lin.17@student.rug.nl
Abstract
Different from other researches which focus on one type of capital flows, one specific episode, or even one country group, this paper aims to study the impact of capital flows on financial stability, while making a comparison between different capital flows, different time episodes, and different country groups. Using a fixed-effects model and studying 34 countries covering the period from 1985 to 2011, we find that the increase of capital inflows tends to raise financial instability. In addition, portfolio flows are more likely to trigger financial instability than other two types of capital flows. Moreover, we study how other factors affect financial stability. These factors are financial development, Total Factor Productivity (TFP), capital account openness, and exchange rate regime. Importantly, we find that 1) the rise of TFP during bad booms can effectively reduce financial instability; 2) to achieve independent monetary policies and financial stability, the management of capital account openness is necessary; 3) adopting a flexible exchange rate regime can help countries to reduce the negative impact of net capital inflows on financial stability.
Table of Contents
Abstract ... 2
1. Introduction ... 4
2. Literature Review... 6
2.1 The determinants of capital flows ... 6
2.2 The influence of capital flows on financial stability ... 7
2.2.1 The influence on economic growth ... 7
2.2.2 The influence on asset prices ... 8
2.2.3 The influence on credit ... 8
3. Data and Methodology ... 9
3.1 Country Sample ... 9
3.2 Baseline Model and Variables ... 9
3.3 Identification of Credit Boom and Bust periods ... 16
4. Empirical Results ... 18
4.1 Baseline Results ... 18
4.2 Further Analyses ... 22
5. Conclusions ... 24
Bibliography: ... 27
Appendix 1. Country information & Time Construction... 32
1. Introduction
Financial crises have a long-lasting and negative impact on economic growth. Due to globalization, the influence of capital flows on financial stability has increased greatly. Nowadays, financial instability happening in a single country, especially a large economy, can be transferred to other countries, which will cause a worldwide financial crisis and an economic recession easily (Cecchetti, Kohler, and Upper, 2009). Therefore, we need to keep an eye on monitoring financial instability and its driving forces.
Since the 1990s, cross-border capital flows have more frequently played an essential role in triggering financial crises. For example, the huge current account deficit combined with large amount of capital inflows in Mexico stimulated domestic bubbles. A following sudden reversal of capital flows sparked the severe financial crisis in 1994 (Carstens and Schwartz, 1998). In 1997, the Asian financial crisis started, which is quite similar to the Mexican crisis. These countries all faced unsustainable euphoric capital inflows before the crash and a run of capital during crisis (Radelet and Sachs, 1998). Although there was a capital contraction during crisis time, the overall trend of international capital flows is expanding. Both the amount and frequency of capital flows have risen a lot.
Even though capital flows described above are linked with financial instability, what we cannot deny is the positive effect of capital flows. When a financial system lacks liquidity, the injected liquidity from other countries could have a positive impact on financial stability. Besides, the capital inflows, especially Foreign Direct Investments (FDI), are accompanied by a transfer of modern production techniques, which helps the sustainable growth of an economy.
Therefore, it is important to study how international capital flows influence financial stability. This paper empirically examines the effect of cross-border capital flows on financial stability in both developed and emerging countries during different time episodes, and also focuses on other factors which may influence financial stability.
There are three contributions in this paper. First, we not only divide capital flows into three components of capital flows, but also identify different time episode, to study the influence of capital flows in different time episode, such as boom, normal, and bust periods. Boom periods are those time periods when there are abnormal credit growth. In construct, bust periods are those time periods when there are abnormal credit decline. The left time periods are normal periods. Second, despite of capital flows, we also find other factors that affect financial stability. These factors are capital account openness, exchange rate regime, financial development, and Total Factor Productivity (TFP). It is important that we also study the interaction effects between some of these factors and capital flows. Finally, we link capital flows and “The Impossible trinity” theory first introduced by Obstfeld and Taylor (1997). This theory discovers that it is impossible for a country to have all three of the following at the same time: 1) A fixed exchange rate regime; 2) Free capital flows; 3) Independent monetary policies. Applying the theory to our explanation, we are able to provide some realistic suggestions to control the influence of capital flows.
2. Literature Review
Since the 1990s when capital flows rose and affected financial stability, the number of studies on capital flows has increased a lot. These studies cover a wide range of topics of capital flows. Only the ones which are most related to financial stability will be presented here.
2.1 The determinants of capital flows
Calvo, Leiderman, and Reinhart (1993) come up the distinction between ‘pull’ and ‘push’ factors. Stable inflation rate, structural reform and increasing productivity are regarded as ‘pull’ factors, which are important domestic factors affecting capital flows. Push factors are related to the world economic situation. The cyclical decline in interest rates and the sharp drop of asset prices in core economies are considered to be ‘push’ factors.
A large amount of studies have focused on ‘push’ factors. They verify that sound global economic circumstances stimulate capital flows, while sluggish global economic growth slows down capital flows or even results in a swing of capital flows. Rey (2015) conducts a VAR analysis, which identifies an important global factor, the monetary policy in the center country. This factor affects leverage of global banks, thus capital flows. This research is consistent with the research of Taylor and Sarno (1997), who study the large portfolio flows from the United States to Latin American and Asian countries during 1988-92, and find that the interest rates in United States is an extremely important determinant of short-term capital flows. Albuquerque, Loayza and Serven (2005) investigate the determinants of FDI, and find that global productivity developments affect global growth, thus increasing FDI. Due to the importance of the banking lending channel in capital flows, some scholars study bank lending only. De Haas and Van Horen (2012) emphasize the heterogeneity of the impact of crises on banks, in terms of distance between banks in home and host countries, and the degree of cooperation with host banks. Bruno and Shin (2013, 2015) analyze the driving force behind banking capital flows, and find that the global leverage cycle of banking is the main driving force.
quality financial institutions are likely to face a capital flight to countries with high-quality financial institutions. Forbes (2010) studies the reasons why foreign investment in the U.S. accounts for a large proportion of total investment, despite its lows returns. On the one side, the highly developed financial system is the core attraction. On the other side, the higher proportion of exposure to the U.S. leads to higher capital flows to the U.S.. However, the relationship between ‘pull’ factors and capital flows is not always linear. Calvo, Izquierdo, and Mejía (2008) find that systemic sudden stops tend to happen in a country with higher financial integration up to a point. Beyond that point, there is a significantly negative relationship. But it is confirmed that weak financial development increases the probability of systemic sudden stops.
Some studies have tried to look into both factors at the same time. Calvo, Leiderman and Reinhart (1996) study capital flows to emerging countries in the 1990s. They find that ‘push’ factors, such as cyclical fluctuations of interest rates, play a bigger role than ‘pull’ factors, such as better financial development and GDP growth, in driving capital inflows into emerging economies. Taylor and Sarno (1997) compare the importance of two sets of determinants. They find that ‘push’ factors have a greater power to explain bond flows. For instance, the decline in US interest rates plays an essential role in driving the short-term portfolio flows to emerging economies. Forbes and Warnock (2012) state that global factors, such as global risk, economic growth and financial contagion, work better than domestic factors in explaining extreme gross capital flows.
2.2 The influence of capital flows on financial stability
2.2.1 The influence on economic growth
rate. This especially hurts those countries with a large proportion of dollarized liabilities. In this case, sustainable growth becomes unsustainable.
2.2.2 The influence on asset prices
Recently, the relationship between capital flows and asset price has attracted a lot of attention. An increasing number of studies focus on this topic.
Capital inflows and house prices have a significantly positive relationship. (Aizenman & Jinjarak, 2009; Jansen, 2003; Kim & Yang, 2011). Moreover, Tillmann (2013) shows that capital inflows are the main explanatory factor of house price movements, while the reaction of house price to capital inflows varies from one country to another due to different monetary policies. Similarly, Sá and Wieladek (2010) highlight that capital inflows play a bigger role than easy monetary policy in pushing house prices and creating a housing price boom followed by a financial crisis. Furthermore, Sá, Towbin, and Wieladek (2014) suggest that capital inflow increase not only house prices, but also available credit to private sector. This relationship is more explicit in countries with advanced securitization.
2.2.3 The influence on credit
Borio and Disyatat (2011) argue that volatile capital flows can cause an unsustainable credit boom, thus leading to a financial crisis. Therefore, there is a close link between capital flows and crisis.
Igan, and Tan (2015) specify that it is the non-FDI capital inflows that boost credit growth and even credit booms, but the increase of household credit is more sensitive to the composition of capital inflows than that of cooperates credit.
Overall, there are a great number of researches focusing on the capital flows either theoretically or empirically. However, most of them focus on a specific episode, or specific countries, especially developing countries. The comparison between different episodes and different types of countries is relatively scarce. The number of studies covering the period of recent sovereign crisis is relatively small. Meanwhile, more precise studies, namely the studies which investigate the relationship between different types of capital flows and other factors, such as capital account openness, Total Factor Productivity (TFP), are needed for a more precise policy response.
3. Data and Methodology
3.1 Country Sample
The selection of countries and time period mainly depends on the data availability. We build up a panel dataset of 34 countries covering the period from 1985 to 2011. Appendix 1 shows the countries and their classification according to development level; 21 of them are advanced economies and 13 of them are emerging economies, so we can make a comparison between different countries.
Since the 1980s, the volatility of financial stability has attracted much attention, especially when a financial crisis of an individual country can easily cause a global financial crisis. Many countries have experienced financial instability in the 20th century, such as Latin American debt crisis in 1980s, Black Monday in 1987— the crash in stock market around the world, United States Savings and Loan crisis from 1989 to 1991, Japanese asset-price-bubble crisis in 1990, Mexico economic crisis from 1994, and the Asian financial crisis from 1997. In the 21st century, the frequency and intensity of crises have increased. In 2001, there was a dot-com bubble bursting. The most well-known and serious crisis is the global financial crisis from 2007-2008. Nowadays, Europe is still on the way of recovery from its sovereign debt crisis. This dataset covers the period from 1985 to 2011, which enables including the most recent financial crises and as many countries as possible.
3.2 Baseline Model and Variables
the credit-to-GDP ratio is likely to trigger a crisis, because this kind of increase is usually accompanied by underestimation of risks and financial bubbles. Overall, the surge of private credit signals financial instability.
Based on these studies, we estimate this basic model:
CTG i,t = 𝛽1+ 𝛽2∗ 𝐹𝐿𝑂𝑊𝑖,𝑡−1+ 𝛽3∗ 𝑋𝑖,𝑡−1+ 𝑣 + 𝜂 + 𝑒𝑖,𝑡 (1)
In this equation, i denotes countries, and t presents years. The dependent variable, CTG, is the credit-to-GDP ratio, which is labeled as private credit by deposit money banks to GDP (%) in the Financial Development and Structure Dataset, whose raw data is from IMF’s International Financial Statistics. The time span of Credit-to-GDP ratio is from 1985 to 2011. FLOW is the net capital inflows, and X is the set of control variables, so we are interested in the parameters 𝛽2 and 𝛽3. After running a
Hausman test, we choose to use a fixed effects model, so 𝑣 is year fixed effects, and 𝜂 is country fixed effects. 𝑒 is the error term. All explanatory variables are lagged by one period, to mitigate reverse causality concerns, namely to make sure that it is net capital inflows that influence domestic credit.
0 50 100 150 200 250 -2,500,000.0 -2,000,000.0 -1,500,000.0 -1,000,000.0 -500,000.0 0.0 500,000.0 1,000,000.0 1,500,000.0 2,000,000.0 2,500,000.0 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 Capital Flows of United Kingdom, 1985-2014
Figure 1. Comparison between Gross capital flows and Net capital inflows Data Source: IMF
Gross capital flows are interesting, but IMF (2011) argues that the non stationarities tend to make modeling difficult. However, net capital inflows, are “both stationary and a natural counterpart to the current account” (Bluedorn, Guajardo, & Topalova, 2011, p.126), and “the heart of the external rebalancing debate” (Duttagupta et al., 2011, p.126). Therefore, this paper follows Igan, and Tan (2015) and uses net capital flows, which is calculated as:
0.0 50.0 100.0 150.0 200.0 250.0 -300,000.0 -200,000.0 -100,000.0 0.0 100,000.0 200,000.0 300,000.0 400,000.0 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 Capital Flows of Spain, 1985-2014
Net capital inflows = CIF-COD = (Financial Direct Investment by foreign investors + Portfolio Investment by foreign investors + Other Investment by foreign investors) - (Financial Direct Investment by domestic investors + Portfolio Investment by domestic investors + Other Investment by domestic investors + Reserves)
As shown in above formula, there are four kinds of capital flows, and it is necessary to do a further research using divided capital flows to see which kind of capital flows has the greatest impact on financial stability. The IMF’s Balance of Payments and International Investment Position Manual Sixth Edition (BPM 6) provides the definitions and highlights the differences between these four capital flows. Direct investment is likely to be a long-term relationship, which is related to control with a significant degree of influence and accompanied by additional contributions such as, technology, marketing, and management. In theory, Portfolio investment is most likely to cause financial instability, because portfolio investment directly provide liquidity and flexibility in financial markets through investing in shares and bonds, and this kind of investment is likely to be short-term investment, with a strong speculation motivation. BPM 6 gives another kind of investment, derivatives investment, which is a special kind of financial investment. This is not presented in the formula due to data availability and its small proportion. Other investment includes transactions other than those included in direct investment, portfolio investment, financial derivatives, and reserve assets. It includes investments in loans offered by banks, which are greatly influenced by interest rates or the exchange rate. So, similar to portfolio investment, this kind of investment is also mainly driven by a strong speculation motivation. Reserve assets are external assets controlled by monetary authorities. The use of reserve assets has the distinct motivation for meeting balance of payments financing needs, for undertaking intervention in exchange markets, and for other related purposes, such as maintaining confidence in the currency and the economy. Because Reserve assets is not driven by markets, we do not look into this item in this paper. All these data are from the IMF’s International Financial Statistics (IFS) data files, and cover the time period from 1985 to 2011.
Similar to Igan, and Tan (2015), control variables here include factors that influence credit growth: real GDP growth rate, deposit interest rates, exchange rate regime index, financial development index, capital account openness, and Total Factor Productivity (TFP) index.
According to Igan, and Tan (2015), the inclusion of interest rates controls for the impact of price. But we use interest rates to study the influence of monetary policies on financial stability. Igan, and Tan (2015) also include inflation rate, but we do not. This is because when we run a multicollinearity test, VIF test, we find inflation rate has highest multicollinearity, and when we delete inflation rate from our model, multicollinearity of every variable is under 10. Despite of IFS dataset, European Central Bank (ECB) offers data for interest rates of European countries after 1999 (Greece, after 2001).
Exchange rate regime index is from Reinhart and Rogoff (2004)’s classification, which is updated to 2010. This index ranges from 1, completely fixed exchange rate regime, to 4, completely floating exchange rate regime. Capital control index controls for openness of an economy. This index was initially introduced by Chinn and Ito (2006) to measure restrictions on cross-border financial transactions reported in the IMF's Annual Report on Exchange Arrangements and Exchange Restrictions. Capital account openness index ranges from 0, completely closed capital account, to 1, totally open capital account. When an economy is opener, it is more likely to have more volatile capital flows, which is related to financial stability. For example, capital flows may have more time and money costs to leave out or inflow into a country with less open capital account. An open capital account allows hot-money to come and go freely, which can lead to financial instability. Besides, “The Impossible trinity” theory also indicates that an open economy and fixed exchange rate regime disable the function of monetary policies, which is likely to cause financial instability when there is huge capital inflows and credit grows rapidly.
According to “The Impossible trinity” theory firstly introduced by Obstfeld and Taylor (1997), it is impossible for a country to have all three of the following at the same time:
1) A fixed exchange rate regime; 2) Free capital flows;
With large amounts of trade or, without a developed financial market to reduce risks of floating exchange rates, a country is inclined to choose option 1) and 2), then its monetary policies may loose power. Theoretically, if the capital account is opener, its monetary policies will have less power, so its financial stability is more likely to be out of control. For example, using monetary policies to maintain a fixed exchange rate, and having large net capital inflows, most of which are driven by speculation motivations, at the same time, countries are likely to experience financial instability. On the other hand, if a country chooses option 2) and 3), then it has independent monetary policies. However, if its capital account is opener, the financial stability is likely to be affected by global financial cycle (Rey, 2015). In an extreme case, if country choose option 1) and 3), then it does not have any, either in or out, capital flow, and this is not the case we discuss in this paper. But, in reality, a country can incompletely open its capital account, allow a certain degree of floating exchange rates, and have a certain impact on monetary policies. In other words, countries have countless choice set. The influence of different choice set is controlled and investigated here.
Figure 2. Financial Development Index Pyramid Source: Svirydzenka (2016)
What is new compared to the model built up by Igan and Tan (2015) is that our model also includes Financial Development Index, which shows the development of financial markets in terms of their depth, access, and efficiency (See Figure 2). Though Sá, Towbin, and Wieladek (2014) argue that securitization may amplify the effect of capital flows on credit growth, we use a more comprehensive index, since the access and efficiency are also related to credit increase. For example, even if a country has a high level of securitization, with low access and efficiency of financial system, securitization looses its ability to boost credit growth. Svirydzenka (2016) builds this new dataset and
Financial Development
Financial Institutions Financial Markets
publicizes the Financial Development Index. The least developed financial market scores 0.00014, and the most developed financial market scores 1 in our data set.
Another control variable, which is new in our model, is the TFP index. This index shows the TFP level at current PPPs (USA=1), and is from the Penn World Table, version 8.11. According to
IMF (2011), TFP is one of the most important factors affecting domestic credit.
Finally, to make a comparison between emerging economies and developed economies, model here also includes an indicator of economic development (Emerging or Developed). The classification follows Nielsen (2011).
3.3 Identification of Credit Boom and Bust periods
Those “extraordinary” periods, such as boom and bust periods, are worth most attention, because a boom is usually followed by a bust, and a bust is linked with recession. If we can find some general rules behind the impact of capital flows on financial stability, then we can know when and how to control capital flows, and what kind of policies the authorities should take.
Igan and Tan (2015) only give the method to identify a boom period. However, Bezemer and Zhang (2014) show how to define three different periods of the credit cycle: boom, bust, and normal periods. Similar to their way to split credit cycle, the identification process of this paper also focuses on the change of credit-to-GDP ratio, denoted as CTG i,t.. Firstly, we can get the trend of
credit-to-GDP ratio via a Hodrick-Prescott filter with a smoothing parameter of 100, and get the standard deviation of the cycle component, denoted as α（CTG i,t）. For a credit boom, we need to identify
both the peak and trough. The peak is when CTG i,t > α（CTG i,t）, where α = 1. The latest preceding
year with both CTG i,t < CTG i,t-1 and CTG i,t < CTG i,t+1 , is defined as the trough. The identification of
bust periods requires similar procedures. The first step, the calculation of trend of credit-to-GDP ratio and its standard deviation, is the same as above. But the identification of peaks and troughs needs a reverse procedure. Before finding the peak, we should find the trough first, which is when CTG i,t < α （CTG i,t） , with α = 1. Then we can find the peak, namely the lasted preceding year with both CTG
i,t > CTG i,t-1 and CTG i,t > CTG i,t+1. Different from Bezemer and Zhang (2014), the identification of
both boom and bust excludes the last year of the period. In other words, we exclude the peak year of
boom and the trough year of bust. This is because the last year of boom or bust is the turning-point year, which can capture the changes of capital flows.
Indonesia
Portugal
Figure 3. Identification of booms and busts Data Source: IMF
Additionally, although the method above successfully captures most abnormal changes, we add another condition to capture several booms and bust, which the above method fails to capture, but are
-10.0 -5.0 0.0 5.0 10.0 15.0 20.0 25.0 0.0 50.0 100.0 150.0 200.0 250.0 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11
private credit to GDP ratio the trend of private credit to GDP ratio
cycle component one standard deviation - one standard deviation -20.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 15.0 20.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 20 09 20 11
private credit to GDP ratio the trend of private credit to GDP ratio
above normal level. Take Indonesia and Portugal: in 1991 the cycle component failed to go above one standard deviation, but in 1989 the growth rate of the credit-to-GDP ratio has arrived 21%, and in 1990 it went up to 34%, which is above normal growth rate in most time periods and countries. Same thing happened to Portugal from 1985 to 1990, and the growth rate in 1986 dropped to -24% already (See Figure 3). For detailed episodes identification, see Appendix.
Following Igan and Tan (2015), we set a threshold, ±20%, for GDP growth rate. Firstly, we find the growth rate of private-credit-to-GDP ratio which is above 20% or under -20%, then find the start and end year of the boom or bust. For boom periods, the start year is the latest preceding year with both CTG i,t < CTG i,t-1 and CTG i,t < CTG i,t+1, the end year is the latest following year with both CTG i,t >
CTG i,t-1 and CTG i,t > CTG i,t+1. For bust periods, the conditions are reverse.
Following Bezemer and Zhang (2014), we also identify good booms and bad booms to see the differences between them. Good booms are those booms not followed by a bust, whereas bad booms are those followed by a bust. Boom periods include both good booms and bad booms.
4. Empirical Results
The summary statistics for all variables are presented in Table 1. We use the fixed-effects model to study the impact of capital flows on financial stability. This section presents all empirical results, where the influence of net capital influence and its three components on financial instability is our main interest. We also focus on other factors that are strongly associated with private credit to GDP ratio in this section, such as financial development index, TFP, capital account openness, and exchange rate regime. Besides, there are further analyses on the interacted influences between these factors and capital flows. Motivated by “the impossible trinity” theory, we investigate how the exchange rate regime affects the influence of net capital inflows on financial stability. The results will be shown at the end.
4.1 Baseline Results
stability. Besides, as capital account openness goes up, private credit tends to increase at 0.01 significance level. There are two possible mechanisms about how capital account openness influences credit and financial stability. In first mechanism, an opener capital account allows capital inflows to push up the price of finance, insurance and real estate (FIRE) sectors more easily. For instance, if the country sets a quota for capital flows, capital may take some time to flow into the country. Without the quota, capital flows freely. As a result, the price of collateral tends to increase rapidly, and more credit can be created. In second mechanism, with an opener capital account, economies are more likely to be affected by the global financial cycle, thus loosing the independence of monetary policies (Rey, 2015). In this case, taking tight monetary policies (raise interest rates) to make an adjustment to excessive credit, the authorities tend to fail, since simultaneously there might be an increasing global liquidity which is likely to flow into countries with higher interest rates. In short, once the monetary policies become ineffective, financial instability goes up. In addition, the increase of financial development index is likely to raise domestic credit at a significance level of 0.01. The value of financial development index shows the depth of finance, which is exploited to indicate financial innovation and securitization. Although these developments increase risk sharing, they allow more credit creation via increasing leverage, which, on the bright sight, can provide credit to market where the demand for credit is above the supply, but, on the dark sight, can increase financial instability.
currently using, is floating. Accordingly, we add a dummy variable with 1 indicating non-Eurozone countries, and interact it with exchange rate regime, to study how the exchange rate regime index affects financial stability in non-Eurozone countries.
Table 3 gives the results after we interact “Non-Eurozone countries” dummy variable with exchange rate regime. In Column (1), (4), (5), and (6), exchange rate regime index has a significantly negative relationship with credit in non-Eurozone countries. This is in line with our expectations. During busts (See Column (6) of Table 3), the negative relationship between exchange rate regime and credit possibility accounts for the fact that some countries, such as Thailand, changed their fixed exchange rate regime to flexible one during busts since they have run out their reserves to against depreciation. To give a clearer presentation of all results, we do not provide the influence of exchange rate regime in non-Eurozone countries for every table. However, we do find that though without “Non-Eurozone countries” dummy variable, exchange rate regime index shows a positive relationship with credit, for non-Eurozone countries, the coefficient of exchange rate regime is significantly negative.
In Column (6) of Table 2, financial development index also shows a significantly positive relationship with credit at 0.01 level, which is in accord with the research by Mehrling (2013). “Moneyness” increases in expansion of credit, which means that lending and financial derivatives are more likely to be treated as money during booms, vice versa. In other words, financial development in terms of financial depth goes well and creates more credit during booms, whereas financial development drops and reduces credit during busts. Therefore, this is a positive cycle during booms, but becomes a negative cycle during busts.
The coefficient of financial development index remains significantly positive. Interestingly, the differences between good booms and bad booms are remarkable according to Column (3) and (4). Even though FDI inflows and other inflows significantly affect financial stability in both good booms and bad booms, net portfolio inflows barely make a difference during good boom periods, but play a significant role during bad boom periods at a significance level of 0.01. Similar to Table 2, the increasing financial development index tends to raise credit during good booms, but does not significantly have an impact on financial stability during bad booms. In contrast, the relationship between capital account openness and credit is insignificant during good booms, but turns into significant during bad booms at 0.01 level. Besides, during bad booms TFP index also presents a negative relationship with credit at 0.05 significance level (See Column (4)). This means that the decline of TFP tends to raise financial instability, or that the increase of TFP is likely to reduce financial instability. Gorton and Ordoñez (2016) recently finish an empirical research about how TFP affects financial stability. They argue that lower TFP reduces the average quality of investment projects. Consequently, the default rate goes up, leading to financial instability. However, in theory, the increase of TFP drives up the credit. One possible explanation for the negative relationship between TFP and credit is that as TFP goes down, the investments flow into FIRE sectors, leading to bad booms. This is might be the reason why the negative relationship between TFP and credit shows significant only during bad booms. In Column (5), it shows that, during normal periods, the rise of both three components of capital inflows and financial development could increase credit. In addition, exchange rate regime index and capital account openness also make a difference during normal periods. During Busts, it is worth noting that among three types of capital inflows, only net portfolio inflows show a significantly positive relationship with credit (See Column (6)). This means that during busts, the decrease of net portfolio inflows is one of the factors leading to credit busts. It could also be inferred that the increase of net portfolio inflows can inhibit credit busts. Similar to the results of Column (6) in Table 2, exchange rate regime index and financial development index can affect credit.
high concentration, the banking system in emerging countries are less efficient to allocate short-term deposits to long-tern loans (Turner, 2006). Accordingly, the increase of net other capital inflows is less effective in emerging markets. This is detrimental to economic growth when credit supply cannot meet the demand in the market, but seems to be beneficial to financial stability when there are excessove net other capital inflows. Interestingly, there is no significant difference between emerging and developed countries in terms of the influence of financial development. Meanwhile, during booms, good booms, and bad bomms, there is no significant difference between emerging and developed countries in terms of the impact of either capital flows or financial development index. (See Column (2) (3) and (4)). The result of normal periods is similar to overall periods (See Column (5)), where there are differences between emerging and developed countries in terms of net other capital inflows. Again, we fail to find significant differences between emerging and developed countries in terms of the influence of either capital flows or financial development during busts (See Column (6)). In addition, The significant positive relationship between portfolio inflows and credit and the significant positive relationship between financial development and credit still hold after we add the market indicator variable.
Finally, we will analyze in depth, TFP index, capital account openness, and exchange rate regime, since these factors may also, to a great extent, affect the influence of net capital inflows. Therefore, the interaction between these factors and capital flows is our main interest. The analyses and further explanations will be given in Section 4.2.
4.2 Further Analyses
All these deeper analyses focus on the interaction effects between other factors and net capital inflows. These factors are TFP index, capital account openness, and exchange rate regime.
beneficial to financial stability during bad booms. This means that the “default” theory that with lower TFP, the default rate of investment projects goes up thus raising financial instability, can be applied to both domestic investments and foreign investments. At the same time, the presence of capital inflows enhances the negative influence of TFP on financial stability. In other words, with rising net capital inflows, the negative impact of TFP also go up. This might account for the fact that foreign investments also flow into FIRE sectors with strong motivation of speculation since decreasing TFP lowers the rate of return of other sectors. As a result, both the decline of TFP and the increase of credit driven by rising price of FIRE sector make the financial system fragile.
Table 7 shows how capital account openness affects the influence of net capital inflows on financial stability, and we mainly focus on three key time periods, boom, normal, and bust periods. As stated above, there are two possible mechanisms about how capital account openness influences credit and financial stability. In first mechanism, an opener capital account allows capital inflows to push up the price of FIRE sectors more easily, for decreasing time costs to flow in or out the economy. As a result, the price of collateral increases, and more credit can be created. In second mechanism, with an opener capital account, economies are more likely to be affected by the global financial cycle, thus loosing the independence of monetary policies (Rey, 2015). Once the monetary policies become ineffective, financial instability goes up. Based on Column (1) and (2) of Table 7, the significantly positive interaction term between net capital inflows and capital account openness during boom and normal periods implies that they have positive moderating effects to each other. In other words, with opener capital account, net capital account is more influential. Meanwhile, with higher capital inflows, capital account openness is also more effective. The significantly positive interaction term between these two factors is accord with above two mechanisms.
Table 8 provides the results of the three-way interaction among the “Non-Eurozone countries” dummy variable, exchange rate regime index, and net capital inflows, and we focus on the impact of exchange rate regime on the influence of net capital inflows on financial stability in non-Eurozone countries. During booms, with all significant interaction terms, one unit increase of net capital inflows to GDP ratio can raise [328.5-68.4*Exchange rate regime index] private credit to GDP ratio (see Column (1)), which means that countries with a more flexible exchange rate regime, the negative impact of net capital inflows on financial stability are smaller during booms. Column (2) does not see a significant interaction term between exchange rate regime index. In contrast, during busts, the influence of net capital inflows significantly depends on exchange rate regime index. One unit decrease of net capital inflows to GDP ratio tends to reduce [44-2.8*Exchange rate regime index] credit, which implies that countries with a more flexible exchange rate regime are less likely to experience a dramatic credit decline. All these results are in line with “The Impossible trinity” theory.
5. Conclusions
In this paper, we study the impact of net capital inflows on financial stability via a fixed-effects model. We build a panel data set including 21 advanced economies and 13 emerging economies and covering a time span from 1985 to 2011. To find precise respondent policies, we divide capital flows into three components of capital flows, and identify different time episode, to study the influence of capital flows in different time episode, such as boom, normal, and bust periods. Moreover, we also find other factors playing a significant role in influencing financial stability. These factors are capital account openness, exchange rate regime, financial development, and Total Factor Productivity (TFP). We further study the interaction effects between some of these factors and capital flows. With these analyses, we are able to provide some realistic suggestions to control the influence of capital flows.
careful about non FDI flows, especially portfolio flows. Among all three kinds of net capital inflows, we only see the differences between emerging and developed countries in terms of net other capital inflows. Compared to the increase of net other capital inflows in developed countries, the increase in emerging markets is less likely to raise credit due to less efficient banking systems in emerging countries.
Other factors also influence financial stability. First, financial development is an important factor affecting financial stability. Due to securitization and some financial innovations via increasing leverage ratio, the increase of financial development index is likely to raise credit. Mehrling (2013) argues that “moneyness” increases in expansion of credit, which means that lending and financial derivatives are more likely to be treated as money during booms, vice versa. In other words, financial development in terms of financial depth goes well during booms, and creates more credit, whereas financial development drops during busts, and reduces credit. Therefore, this is a positive cycle during booms, but becomes a negative cycle during busts. Further, given one unit increase of financial development, emerging countries and developed countries do not show significant differences in terms of credit growth.
Second, during bad booms, TFP shows a negative relationship with credit. The increase of TFP can raise the quality of investment projects, whereas the decrease of TFP is more likely to increase the default rate of companies, thus raising financial instability. Moreover, decreasing TFP makes FIRE sectors more attractive to not only domestic investments but also foreign investments, namely foreign capital inflows, which is likely to cause a credit boom followed by a bust. We find that the increase of TFP during bad booms can be one of the effective methods to control both financial stability and the negative impact of net capital inflows on financial stability.
as “hot money”, and if necessary they can set some restrictions, such as a quota, for these capital flows.
Next, exchange rate regime index also plays a important part in affecting financial stability. We mainly focus on the effects of exchange rate in non-Eurozone countries, since Eurozone is a special case. Eurozone countries have an fixed exchange rate between their original currency and Euro, but have a floating exchange rate between Euro, the currency they are currently using, and other currencies. In general, net capital inflows are more influencial during both boom and bust periods in countries with a more fixed exchange rate regime, which is consistent with “The Impossible trinity” theory first introduced by Obstfeld and Taylor (1997). Therefore, to enhance financial stability, it is better to adopt a floating exchange rate regime.
Bibliography:
Aghion, P., Bacchetta, P., & Banerjee, A. (2004). Financial development and the instability of open economies. Journal of Monetary Economics, 51(6), 1077-1106.
Aizenman, J., & Jinjarak, Y. (2009). Current account patterns and national real estate markets. Journal of Urban Economics, 66(2), 75-89.
Albuquerque, R., Loayza, N., & Servén, L. (2005). World market integration through the lens of foreign direct investors. Journal of International Economics, 66(2), 267-295..
Bezemer, D. J., & Zhang, L. (2014). From boom to bust in the credit cycle: the role of mortgage credit. University of Groningen, Faculty of Economics and Business, 14025-GEM.
Borio, C. E., & Disyatat, P. (2011). Global imbalances and the financial crisis: Link or no link?. Bank for International Settlements Working Papers 346.
Bosworth, B. P., Collins, S. M., & Reinhart, C. M. (1999). Capital flows to developing economies: implications for saving and investment. Brookings papers on economic activity, 1999(1), 143-180. Broner, F., Didier, T., Erce, A., & Schmukler, S. L. (2013). Gross capital flows: Dynamics and crises. Journal of Monetary Economics, 60(1), 113-133.
Bruno, V., & Shin, H. S. (2013). Capital flows, cross-border banking and global liquidity . National Bureau of Economic Research (No. w19038).
Bruno, V., & Shin, H. S. (2015). Capital flows and the risk-taking channel of monetary policy. Journal of Monetary Economics, 71, 119-132.
Caballero, R. J., & Krishnamurthy, A. (2006). Bubbles and capital flow volatility: Causes and risk management. Journal of monetary Economics,53(1), 35-53.
Calvo, G. A. (1998). Capital flows and capital-market crises: the simple economics of sudden stops. Journal of Applied Economics, Universidad del CEMA, November, 35-54.
Calvo, G. A., Izquierdo, A., & Talvi, E. (2003). Sudden stops, the real exchange rate, and fiscal sustainability: Argentina's lessons. National Bureau of Economic Research No. w9828.
Calvo, G. A., Izquierdo, A., & Mejía, L. F. (2008). Systemic sudden stops: the relevance of balance-sheet effects and financial integration. National Bureau of Economic Research No. w14026.
Carstens, A., & Schwartz, M. J. (1998). Capital flows and the financial crisis in Mexico. Journal of Asian Economics, 9(2), 207-226..
Cecchetti, S. G., Kohler, M., & Upper, C. (2009). Financial crises and economic activity. National Bureau of Economic Research No. w15379.
Chinn, M. D., & Ito, H. (2006). What matters for financial development? Capital controls, institutions, and interactions. Journal of development economics, 81(1), 163-192.
Dabla-Norris, E., Ho, G., Kochhar, K., Kyobe, A., & Tchaidze, R. (2014). Anchoring Growth: The Importance of Productivity-Enhancing Reforms in Emerging Market and Developing Economies. Journal of International Commerce, Economics and Policy, 5(02), 1450001.
De Haas, R., & Van Horen, N. (2012). Running for the exit? International bank lending during a financial crisis. Review of Financial Studies, hhs113..
Devigne, D., Vanacker, T., Manigart, S., & Paeleman, I. (2013). The role of domestic and cross-border venture capital investors in the growth of portfolio companies. Small Business Economics, 40(3), 553-573.
Duttagupta, R., Bluedorn, J., Guajardo, J., & Topalova, P. (2011). International capital flows: reliable or fickle. Charpter 4, IMF World Economic Outlook 2011, 125-63.
Erel, I., Liao, R. C., & Weisbach, M. S. (2012). Determinants of cross‐border mergers and acquisitions. The Journal of Finance, 67(3), 1045-1082.
Forbes, K. J. (2010). Why do foreigners invest in the United States?. Journal of International Economics, 80(1), 3-21.
Forbes, K. J., & Warnock, F. E. (2012). Capital flow waves: Surges, stops, flight, and retrenchment. Journal of International Economics, 88(2), 235-251.
Gourinchas, P. O., & Jeanne, O. (2006). The elusive gains from international financial integration. The Review of Economic Studies, 73(3), 715-741.
Hernández, L., & Landerretche, O. (2002). Capital inflows, credit booms, and macroeconomic vulnerability: the cross-country experience. Banking, Financial Integration, and International Crises, Central Bank of Chile Santiago, Chile, 199-233.
Houben, A., R. van der Molen and P. Wierts (2012), Making macroprudential policy operational, Revue de Stabilité Financière, Banque Centrale du Luxembourg, 13-25.
Igan, D., & Tan, Z. (2015). Capital Inflows, Credit Growth, and Financial Systems. International Monetary Fund No. 15-193.
Ivashina, V., & Scharfstein, D. (2010). Bank lending during the financial crisis of 2008. Journal of Financial economics, 97(3), 319-338.
IMF (2011). World Economic Outlook. International Monetary Fund.
Jansen, W. J. (2003). What do capital inflows do? Dissecting the transmission mechanism for Thailand, 1980–1996. Journal of Macroeconomics, 25(4), 457-480.
Ju, J., & Wei, S. J. (2011). When is quality of financial system a source of comparative advantage?. Journal of International Economics, 84(2), 178-187.
Kim, S., & Yang, D. Y. (2011). The impact of capital inflows on asset prices in emerging Asian economies: is too much money chasing too little good?.Open Economies Review, 22(2), 293-315. Lane, P. R., & McQuade, P. (2014). Domestic Credit Growth and International Capital Flows*. The Scandinavian Journal of Economics,116(1), 218-252.
Mendoza, E. G., & Terrones, M. E. (2008). An anatomy of credit booms: evidence from macro aggregates and micro data. National Bureau of Economic Research No. w14049.
Mehrling, P. (2013). 21 The inherent hierarchy of money. Social Fairness and Economics: economic essays in the spirit of Duncan Foley, 169, 394.
Mihaljek, D. (2008). The financial stability implications of increased capital flows for emerging market economies. BIS papers, 44, 11-44.
Nkusu, M. (2011). Nonperforming loans and macrofinancial vulnerabilities in advanced economies. IMF Working Papers No. 11/161, 1-27.
Nielsen, L. (2011). Classifications of countries based on their level of development: How it is done and how it could be done. IMF Working Papers No. 11/31, 1-45.
Obstfeld, M., & Taylor, A. M. (1997). The great depression as a watershed: international capital mobility over the long run (No. w5960). National Bureau of Economic Research.
Ordonez, G., & Gorton, G. (2015). Good Booms, Bad Booms (No. 292). Society for Economic Dynamics, 2015 Meeting Papers 292.
Powell, A., Ratha, D., & Mohapatra, S. (2002). Capital inflows and outflows: Measurement, determinants, consequences. Unpublished paper, University of Torcuato Di Tella. Available at http://www. utdt. edu/Upload/CIFPwp/wpcif $072002. pdf.
Radelet, S., & Sachs, J. (2000). The onset of the East Asian financial crisis. In Currency crises, University of Chicago Press, 105-153.
Rey, H. (2015). Dilemma not trilemma: the global financial cycle and monetary policy independence . National Bureau of Economic Research No. w21162.
Sá, F., & Wieladek, T. (2010). Monetary policy, capital inflows and the housing boom. Bank of England Working Paper No. 405.
Taylor, M. P., & Sarno, L. (1997). Capital flows to developing countries: long-and short-term determinants. The World Bank Economic Review, 11(3), 451-470.
Tillmann, P. (2013). Capital inflows and asset prices: Evidence from emerging Asia. Journal of Banking & Finance, 37(3), 717-729.
Tong, H., Bakker, B., & Vandenbussche, J. (2012). Policies for macrofinancial stability: How to deal with credit booms. IMF Staff Discussion Notes No. 12/6.
Turner, P. (2006, August). The banking system in emerging economies: how much progress has been made?. BIS Papers No 28.
Appendix 1. Country information & Time Construction
1. Country group
Country Market Classification Country Market Classification
Australia Developed Japan Developed
Barbados Emerging Korea, Republic of Developed
Brazil Emerging Mexico Emerging
Canada Developed Netherlands Developed
China, P.R. Emerging Norway Developed
Costa Rica Emerging Portugal Developed
Cyprus Developed Russian Federation Emerging
Denmark Developed Seychelles Emerging
Finland Developed Singapore Developed
France Developed South Africa Emerging
Germany Developed Spain Developed
Greece Developed Swaziland Emerging
Hungary Emerging Sweden Developed
India Emerging Switzerland Developed
Indonesia Emerging Thailand Emerging
Israel Developed United Kingdom Developed
2. Time Construction
Country Time Span Episode Boom Bust Normal
Appendix 2. Tables
Table 1. Summary Statistics for Key Variables
Variables Mean Std Min Max
Private Credit to GDP ratio 74.23123 47.60999 6.33384 284.62180
Net Capital Inflows to GDP ratio -0.00261 0.06629 -0.27166 0.80785
Net FDI Inflows to GDP ratio 0.00770 0.03800 -0.20387 0.37520 Net Portfolio Inflows to GDP ratio -0.00001 0.08139 -0.59982 1.47422
Net Other Inflows to GDP ratio 0.00251 0.06511 -0.64577 0.84061 Real GDP Growth Rate 3.22921 3.59682 -14.53107 21.01800
Deposit Interest rates 35.75645 409.13610 0.03000 9394.2930 Exchange Rate Regime Index 2.64351 1.19242 1.00000 4.00000
Capital Account Openness 0.67050 0.35876 0.00000 1.00000 Financial Development Index 0.51368 0.23219 0.00014 1.00000
Table 2. Baseline Result 1
(1) (2) (3) (4) (5) (6)
Overall Boom Good Boom Bad Boom Normal Bust Net Capital Inflows to GDP ratio 111.8*** 117.4*** 97.53* 310.5*** 124.0*** 77.36
(21.14) (42.40) (48.95) (66.33) (21.57) (51.75) Real GDP Growth Rate -1.689*** -0.410 -0.593 0.130 -1.506*** -3.701***
(0.306) (0.513) (0.543) (1.227) (0.298) (0.725) Deposit Interest Rate 0.000307 -0.000660 0.00115 -0.0726 0.00162 -0.00313
(0.00184) (0.00207) (0.00224) (0.0770) (0.00444) (0.00498) Exchange Rate Regime Index -0.335 -2.788 -3.101 6.606* 3.702*** 4.844**
(1.028) (1.962) (3.310) (3.717) (1.234) (2.422) Capital Account Openness 14.57*** 6.223 37.52 58.24*** 24.55*** 18.19
(5.213) (11.81) (22.82) (16.03) (5.176) (12.43) Financial Development Index 114.3*** 94.65*** 123.2*** 0.899 123.3*** 61.19***
(6.852) (14.52) (27.02) (15.03) (6.972) (19.09) TFP Index 11.41 1.903 -19.49 -66.92 8.839 -16.33 (10.60) (19.75) (25.53) (53.02) (10.65) (36.86) Constant 7.113 35.03* 16.46 90.89* -17.08* 51.13 (9.479) (18.00) (27.70) (49.90) (10.20) (31.67) Observations 679 178 95 83 355 146 R-squared 0.435 0.423 0.382 0.524 0.623 0.334 Number of id 33 30 20 23 33 27
Country FE YES YES YES YES YES YES
Year FE YES YES YES YES YES YES
Table 3. Baseline Result 1.1
(1) (2) (3) (4) (5) (6)
Overall Boom Good Boom Bad Boom Normal Bust Net Capital Inflows to GDP ratio 110.3*** 121.3*** 108.8** 260.7*** 132.2*** 24.82
(20.88) (42.23) (49.47) (65.08) (21.48) (48.66) Real GDP Growth Rate -1.684*** -0.432 -0.657 0.307 -1.584*** -2.895***
(0.302) (0.511) (0.543) (1.158) (0.296) (0.685) Deposit Interest Rate 0.000250 -0.000535 0.00136 -0.0690 0.00162 -0.00189
(0.00182) (0.00206) (0.00223) (0.0726) (0.00439) (0.00457) Exchange Rate Regime Index 5.630*** 0.827 -0.644 9.845** 8.194*** 22.20*** (1.760) (2.981) (3.800) (3.696) (1.941) (4.274) NonEuro × Exchange Rate
Regime Index
-8.971*** -6.257 -8.634 -32.11*** -8.794*** -26.03*** (2.162) (3.900) (6.662) (11.63) (2.621) (5.480) Capital Account Openness 14.62*** 5.442 34.69 53.13*** 24.31*** 21.97* (5.148) (11.76) (22.81) (15.23) (5.113) (11.41) Financial Development Index 116.8*** 96.33*** 127.8*** 1.390 130.6*** 55.26***
(6.794) (14.48) (27.11) (14.17) (7.313) (17.53) TFP Index 15.57 5.851 -20.44 -78.92 11.90 38.47 (10.52) (19.79) (25.42) (50.18) (10.57) (35.67) Constant 5.711 34.02* 26.08 154.3*** -17.56* 14.63 (9.367) (17.91) (28.54) (52.35) (10.08) (30.00) Observations 679 178 95 83 355 146 R-squared 0.449 0.433 0.397 0.585 0.633 0.447 Number of id 33 30 20 23 33 27
Country FE YES YES YES YES YES YES
Year FE YES YES YES YES YES YES
Table 4. Baseline Result 2
(1) (2) (3) (4) (5) (6)
Overall Boom Good Boom Bad Boom Normal Bust Net FDI Inflows to GDP ratio 55.62 32.05 197.1* 340.9*** 113.3*** 28.94
(37.72) (65.89) (108.5) (82.89) (37.09) (89.21) Net Portfolio Inflows to GDP ratio 49.16** 5.983 -38.22 364.2*** 53.67** 103.5* (24.43) (51.27) (61.90) (87.73) (24.22) (54.45) Net Other Inflows to GDP ratio 109.0*** 84.56* 96.42* 379.9*** 138.1*** 26.43
(21.09) (43.55) (52.85) (75.08) (21.54) (51.97) Real GDP Growth Rate -1.796*** -0.416 -0.558 0.355 -1.623*** -3.313***
(0.306) (0.510) (0.488) (1.212) (0.290) (0.718) Deposit Interest Rate 7.53e-05 -0.000866 0.00106 -0.0724 0.00105 -0.00309
(0.00184) (0.00206) (0.00202) (0.0761) (0.00431) (0.00501) Exchange Rate Regime Index -0.241 -1.127 0.526 2.217 3.318*** 4.859*
(1.031) (2.073) (3.259) (4.372) (1.206) (2.461) Capital Account Openness 11.07** 1.610 32.82 62.88*** 18.68*** 16.69
(5.276) (11.92) (20.87) (16.27) (5.158) (12.53) Financial Development Index 112.5*** 96.88*** 117.6*** 0.165 120.1*** 69.88***
(6.836) (14.53) (24.60) (15.17) (6.799) (19.25) TFP Index 5.050 -8.493 -35.20 -157.2*** -0.369 -14.36 (10.47) (20.02) (24.66) (53.90) (10.26) (37.54) Constant 14.77 40.24** 21.04 168.5*** -3.631 43.85 (9.461) (18.28) (27.68) (50.88) (9.989) (32.01) Observations 679 177 95 82 354 148 R-squared 0.437 0.441 0.513 0.562 0.649 0.336 Number of id 33 30 20 23 33 28
Country FE YES YES YES YES YES YES
Year FE YES YES YES YES YES YES
Table 5. Comparison between Developed and Emerging market
(1) (2) (3) (4) (5) (6)
Overall Boom Good Boom Bad Boom Normal Bust Net FDI Inflows to GDP ratio 66.91 35.43 299.1* 352.9*** 72.78* 40.03
(43.11) (89.29) (165.9) (101.8) (42.46) (99.14) Emerging Market × Net FDI Inflows
to GDP ratio
20.33 71.12 -134.5 -691.2 87.50 -71.63 (93.80) (177.8) (231.0) (915.4) (85.56) (245.8) Net Portfolio Inflows to GDP ratio 79.88*** 18.47 6.094 375.4*** 69.86*** 121.5** (28.75) (74.29) (93.31) (106.6) (26.50) (58.88) Emerging Market × Net Portfolio
Inflows to GDP ratio
45.51 -5.073 0.682 290.4 69.47 1.981 (83.61) (158.9) (173.8) (530.6) (72.29) (255.0) Net Other Inflows to GDP ratio 150.3*** 97.76 149.3* 393.9*** 174.6*** 61.14
(26.58) (66.31) (85.96) (101.3) (24.69) (60.71) Emerging Market × Net Other
Inflows to GDP ratio
-122.2*** -28.30 -87.14 19.25 -158.7*** -151.3 (46.20) (97.21) (108.8) (195.1) (47.84) (122.6) Real GDP Growth Rate -1.718*** -0.381 -0.477 0.331 -1.666*** -3.465***
(0.308) (0.527) (0.526) (1.301) (0.289) (0.746) Deposit Interest Rate -0.000189 -0.000836 0.000699 -0.0673 0.000354 -0.00513
(0.00185) (0.00210) (0.00209) (0.0815) (0.00430) (0.00553) Exchange Rate Regime Index -0.121 -1.236 0.585 2.476 3.968*** 4.843*
(1.027) (2.133) (3.503) (4.643) (1.194) (2.485) Capital Account Openness 10.62** 1.343 32.82 66.89*** 17.80*** 16.43
(5.258) (12.69) (21.34) (21.43) (5.193) (12.68) Financial Development Index 115.6*** 98.32*** 134.3*** -1.879 120.7*** 63.89***
(7.395) (16.73) (36.46) (16.58) (7.626) (20.66) Emerging Market × Financial
Development Index -10.77 -8.516 -30.54 -5.867 9.028 63.35 (17.33) (30.32) (52.98) (80.87) (16.28) (61.36) TFP Index 6.654 -10.45 -36.28 -141.3* -6.901 -4.830 (10.54) (21.72) (29.38) (82.48) (10.37) (38.70) Constant 11.97 41.57** 15.94 153.8** -1.483 34.67 (9.544) (20.12) (30.24) (75.62) (9.841) (33.05) Observations 679 177 95 82 354 148 R-squared 0.447 0.444 0.523 0.569 0.669 0.350 Number of id 33 30 20 23 33 28
Country FE YES YES YES YES YES YES
Year FE YES YES YES YES YES YES
Table 6. Interaction between capital flows and TFP
Bad Boom Net Capital Inflows to GDP ratio 803.8***
(294.7)
TFP Index -117.3*
(59.78) Net Capital Inflows to GDP ratio×TFP Index -644.8* (375.6)
Real GDP Growth Rate 0.330
(1.210)
Deposit Interest Rate -0.0831
(0.0759) Exchange Rate Regime Index 5.491
(3.708) Capital Account Openness 45.30**
(17.46) Financial Development Index 8.941
(15.48) Constant 138.7** (56.38) Observations 83 Number of id 23 R-squared 0.550 Country FE YES Year FE YES
Table 7. Interaction between capital flows and capital account openness
(1) (2) (3)
Boom Normal Bust
Net Capital Inflows to GDP ratio 17.55 29.14 -13.65 (66.84) (37.33) (102.1) Capital Account Openness 7.805 19.78*** 18.23
(11.73) (5.334) (12.43) Net Capital Inflows to GDP ratio × Capital
Account Openness
202.1* 151.0*** 144.8 (105.2) (48.83) (140.1) Real GDP Growth Rate -0.225 -1.446*** -3.654***
(0.518) (0.295) (0.726) Deposit Interest Rate -0.000618 0.000493 -0.00308
(0.00205) (0.00440) (0.00498) Exchange Rate Regime Index -2.938 3.518*** 4.856**
(1.945) (1.219) (2.421) Financial Development Index 94.42*** 126.9*** 63.88***
(14.38) (6.974) (19.26) TFP Index 4.462 7.612 -18.32 (19.61) (10.52) (36.90) Constant 31.59* -13.69 51.14 (17.92) (10.13) (31.66) Observations 178 355 146 R-squared 0.437 0.634 0.341 Number of id 30 33 27
Country FE YES YES YES
Year FE YES YES YES
Table 8. Three-way interaction among Eurozone countries indicator, capital flows, and exchange rate regime
(1) (2) (3)
Boom Normal Bust
Net Capital Inflows to GDP ratio 328.5*** 276.1*** -1,080*** (123.0) (90.60) (214.0) NonEuro × Net Capital Inflows to GDP ratio -117.3 -276.8** 1,124***
(160.2) (108.6) (227.8)
Exchange Rate Regime Index -4.830 6.982*** 8.107
(3.590) (1.889) (5.241) NonEuro × Exchange Rate Regime Index -1.885 -7.581** -12.81**
(4.410) (2.565) (5.993) Net Capital Inflows to GDP ratio × Exchange Rate Regime
Index
136.4* 50.87 470.5*** (71.72) (53.92) (83.29) NonEuro × Net Capital Inflows to GDP ratio × Exchange
Rate Regime Index
-204.8** -21.80 -473.3*** (80.19) (56.95) (87.98)
Real GDP Growth Rate -0.554 -1.344*** -2.117***
(0.473) (0.289) (0.623)
Deposit Interest Rate -0.000887 0.00109 -0.00107
(0.00190) (0.00423) (0.00404)
Capital Account Openness 5.033 22.11*** 23.75**
(11.26) (5.024) (10.23) Financial Development Index 99.06*** 129.2*** 50.00***
(13.84) (7.091) (15.53) TFP Index 6.022 7.567 33.20 (18.25) (10.35) (31.64) Constant 34.39** -14.23 25.35 (16.49) (9.814) (26.67) Observations 178 355 146 R-squared 0.531 0.662 0.580 Number of id 30 33 27
Country FE YES YES YES
Year FE YES YES YES
