CC BY
48What Impact Do Currency Exchange Rates Have on the M&A Market in BRICS Countries?
Kristina Bondareva
M.A., Department of corporate finance and corporate governance, Finance University under the Government of the Russian Federation
Abstract. This paper tries to examine how currency exchange rates are influencing the M&A market in BRICS countries. Therefore the amount of M&A deals is defined as the dependent variable. Next to the currency exchange rate further variables like GDP growth rate, Stock (size of stockmarket) and money and quasi money growth are included this model. This data was gathered by the World Bank and modifyed for the right purpose. We used yearly data from 1994-2014 by 4 different countries. But in consequence of the fact that not all the data is availiable since 1994 we were able to obtain 64 observations. By using panel data with fix effects and lags this paper tries to display the impact of currency exchange rates on the M&A market through 4 cross-sectional units in a time period of 14 years (without timelags). After estimating the model we came to the conclusion that currency exchanges have a negative effect which is mostly sicnificant in the second period.
Keywords: M&A, BRICS countries, exchange rates, panel data model, fixed effect estimator, lags.
Влияние обменных курсов на рынок слияний и поглощений в странах БРИКС
Кристина Бондарева
магистр, Департамент корпоративных финансов и корпоративного управления, Финансовый
университет, Москва, Россия
Аннотация. В данной статье исследуется влияние обменных курсов валют на рынок слияний и поглощений в странах БРИКС. Таким образом, количество сделок M&A задается в качестве зависимой переменной. Наряду с обменными курсами валют, анализируются такие переменные, как темпы роста ВВП, размер фондового рынка, темпы роста денежной массы. Данные для исследования были собраны с сайта Всемирного банка и модифицированы для целей регрессионной модели. Ввиду неполноты информации за период 1994-2014 гг. для 4 стран удалось найти 64 наблюдения. Использование временных рядов с фиксированными эффектами и временными лагами позволило продемонстрировать влияние валютных курсов на M&A через 4 cross-sectional выборки на временном интервале в 14 лет (без учета временных лагов). По результатам оценки модели можно сделать вывод о том, что рост обменного курса имеет негативное влияние на совершение сделок M&A, наиболее значимо данный эффект проявляется в следующем периоде.
Ключевые слова: слияния и поглощения; страны БРИКС; валютные курсы; панельные данные; фиксированный эффект; временные лаги.
1. INTRODUCTION
Nowadays M&A represent a significant part of FDI (Foreign Direct Investment). These capital flows have a big impact on the development of countries' economies and their GDP (Gross domestic product) growth (Neto, Brandao & Cer-queira, 2010). Especially, M&A (mergers and acquisitions) could be important "economic driver" for BRICS countries (Brasil, Russia, India & China), which are on the stage of newly advanced economic development. Along with the rest of the world the BRICS countries experience rather high economic volatility, especially in terms of currency exchange rates. For this reason investigating the impact of currency exchange rates on M&A market in BRICS countries is of high interest.
For the aim of our research we gathered mac-roeconomic data for 4 countries from World Bank Database and modified it in cross-sectional units with 14 time periods. To estimate the model we apllied fixed effects tecnique and intriduced lags in order to take into account long-term effects.
Specification of the model is based on Literature review section (2). To specify the model we introduced other related variables and estimated it through fixed effects tecnique of panel data, what is going to be explained in Model section (4). All the data gathered for the observations is described in Data section (3). In section of Emperical results (5) all the estimations could be find.
2. LITERATURE REVIEW
In order to specify the model our first step was to analyze works already done on this or similar topics. The first author to whom we have addressed was Mileva. In her work Mileva (2008) emphasizes that few of studies focus directly on M&A flows. Usually authors consider the total amount of investment flows. It increased our interest in investigating M&A market. Estimating the effect of FDI on domestic investment Mileva based on emerging and transition economies rather than on developed countries. The author said that from long-term perspective each dollar of FDI usually generated at least one additional dollar of local investment. But in less developed countries the effect could differ significantly, what is interesting to study. In our project we decided to stand by this idea and to focus on BRICS countries.
Wong (2008) tried to apply gravity model to explain M&A flows. The investigation showed that geographic, linguistic and colonial variables are not suitable. That is why we decided not to include such variables in our model.
The study of Neto, Brandao and Cerqueira (2010) identifies macroeconomic factors, affecting cross-border M&A. The authors found out that one of the important factors is the size of economy. In our model we have included economic growth (as annual% of GDP growth). Another significant factor is the size of capital markets. For capturing capitalization factor in our model we decided to use the total value of shares traded (as % of GDP).
Hyun and Kim (2010) determining factors of cross-border M&A focused on the role of institutions and financial development. The authors based on gravity model but extended it with some extra variables. For example, applying method of Di Giovanni (2005), who found using panel dataset of M&A that deep financial markets can play a significant role for M&A, Hyun and Kim included in their model financial market development indicators (the stock market capitalization and the amount of credit provided by banks and other financial institutions to the private sector). The authors also supposed that currency exchange rates could affect M&A flows. So, depreciation of the currency can make it more attractive to invest in this country, for example because of decreasing production costs or decreasing value of assets. The estimation of the model showed that market size had positive and significant effect, while coefficient for exchange rates appeared statistically insignificant.
Brooks, Edison, Kumar and Sl0k (2004) also claimed that there is no clear connection between M&A and exchange rates. Authors provided some reasons. First of all lots of cross-border deals are financed through share-swaps. Furthermore, acquiring companies can already have cash in currency or they can issue a debt in that currency. In their model authors investigated the influence of M&A flows on exchange rates and they found the coefficients statistically insignificant. Still we were interested in testing the opposite influence (effect of changes in exchange rates on M&A), including also long-term effects (lags).
Baker, Foley and Wurgler (2009) in their work empirically evaluated the effect of cheap assets on
Table 1. The numbers of M&A deals in BRICS countries since year 1994
Year Brazil Russia India China
1994 97 85 - 106
1995 153 202 - 120
1996 191 163 - 191
1997 233 112 - 302
1998 387 96 - 357
1999 353 210 423 340
2000 530 418 895 530
2001 408 398 721 570
2002 258 403 599 1064
2003 212 501 723 1704
2004 270 406 790 2400
2005 273 477 1283 1951
2006 377 699 1524 2212
2007 871 999 1570 2963
2008 940 1783 1503 3408
2009 530 3357 1372 3089
2010 712 3775 1451 3721
2011 864 3312 1116 4103
2012 836 2610 1169 3810
2013 629 2096 1022 3964
2014 566 1958 1155 5122
Source: https://imaa-institute.org/statistics-mergers-acquisitions/.
FDI. The results didn't support the existence of a cheap asset effect. But we suppose that focusing exactly on M&A deals, which nowadays represent a big part of FDI, can allow us to find a correlation between the costs of assets, what in our model is expressed by changes in currency exchange rates, and the investment flows.
3. DATA
For analyzing M&A market we decided to use the annual numbers of M&A deals. This data was gathered from the IMAA (Institute for Mergers, Acquisitions and Alliances). The web-site allows downloading database for each country. So, we exported to Excel the numbers of M&A deals in Brazil, Russia, India and China (Table 1).
Graphically this information could be presented as did in graph below (Graph 1).
To explain M&A we collected data for exchange rates, annual GDP growth rates, total values of stocks traded and growth rates of amount of money and quasi money in the economies. To gather the statistics we used the World Bank Database. It is possible to export all the data from the web-site to Excel. We used yearly data from 1994 to 2014 for 4 different countries. But in consequence of the fact that not all the data is availiable since 1994 we were able to obtain 64 observations. The results are presented in the annex (Annex 1). For the aims of our project we present exchange rates as differences of logarithms of exchange rates. All variables will be explained in more details later in the section 4.
6 000
5 000
4 000
3 000
2 000
1 000
0
■Brazil
Russia
India
■ China
Graph 1. Amount of M&A deals since 1994
4. MODEL
4.1. Description of variables
4.1.1. Dependent variable
To measure the effect of currency exchanges on M&A we firstly thought about two different ways to model this variable. One way to measure M&A deals is the volume of money (e.g. € in one year). The problem in this case is that one big merger or acquisition can have a huge impact on the data in one year. This distortion can be reduced by describing the dependent variable as the amount of M&A deals in one year. In this case the problem might be that a "small" M&A deal is weighted equally as a "big" deal. But we decided that this way is the most appropriate to describe our dependent variable as it reflects the activism. To specify our model we decided to concentrate just on a few countries because otherwise we would have a very complex model in which it is almost not possible to find any potential relationships. Following the work of Mileva we would like to focus on emerging and transition economies which are rather uniform and experience volatility of currency exchange rates. The BRICS countries fulfill these conditions. Therefore we decided to define the dependent variable as the amount of M&A deals in each of BRICS countries during one year.
As this variable has positive and rather volatile values it is more suitable to apply logarithms.
4.1.2. Independent variables
In the introduction we explained that we are going to analyze the impact of the currency exchange rates on M&A deals in BRICS countries. Except of
the independent variable for currency exchange rates we additionally added in our model other variables like GDP growth rate, size of stockmarket within a country and money and quasi money growth to make the model closer to reality and more statistically significant. So, we gathered data for each of the BRICS countries for our model.
Currency exchange rates
It is a matter of common knowledge that BRICS countries do not use the same currency. Therefore to obtain data in a useful and reasonable form we downloaded the annually exchange rates which were calculated as an annual average (based on monthly averages) of local currency units relative to the U.S. dollar. To make exchange rate of each country comparable we decided to use differences of logarithms for current and previous years. In comparison with actual differences we can now use the percentage differences of the exchange rate in each country as a comparable structure for each country. The problem with actual differences is that they are depending on the quantitative differences of each exchange rate. Because of this we assume that for our purpose the best way to describe and model our first independent variable is as following: Ex = [ln(ext) - ln(ext-1)].
GDP growth rate
In our case the GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. We included the GDP because it is a common and frequently used indicatior not only for macroeconomic purposes but also for financial analysts and investors all over the world. It is used to gauge the health of economy, so investors are concered about negative GDP
Model 7: Fixed-effects, using 56 observations Included 4 cros3-3ectional units Tirce-3erie3 length = 14
Dependent variable: 1 Aicountof 1-IAdea 1 sinnumbers Orr.itted due to exact collinearity: dt
coefficient std. error t-ratio
p-value
const 7, .31450 0. .237533 32 . . 90 6. . 4 3e-02 3 A- A- A-
LNexch I o .562755 0. .779030 0. .7223 0. .4749 I
Mo ne yandqua s iicon~ -0 .0324532 0. .0101957 -3. .133 0 .0031 A- A- A-
Stock3txadedtota~ 0 .00133343 0. .00223337 0. .5343 0 .5625
GDF Gr owthrat e -0 .0344396 0. .0265574 -1. .297 0 .2032
dt 1 -0 .342390 0. .251090 -3. .357 0 .0019 A- A- A-
dt 2 -0 .373937 0. .250126 -3. .514 0 .0012 A- A- A-
dt 3 -0 .431261 0. .263090 -1. .329 0 .0759
dt i -0 .337413 0. .263654 -1. .469 0 .1507
dt 5 -0 .222957 0. .270553 -0. .3241 0 .4155
dt 6 0 .141544 0. .272554 0. .5193 0 .6063
dt 7 0 .476963 0. .295505 1. .614 0 .1155
dt 3 0 .336792 0. .246403 1. .367 0 .1304
dt 9 0 .165914 0. .256141 0. .6477 0 .5214
dt 10 0 .613152 0. .262437 2. .336 0 .0253 fr fr
dt 11 0 .490129 0. .245274 1. . 993 0 .0535 fr
dt 12 0 .153963 0. .231333 0. .6370 0 .4966
dt 13 0 .0257201 0. .231547 0. .1111 0 . 9122
Mean dependent var 7.021701
Sum squared re3id 3.663354
LSDV R—squared 0.904161
LSDV F(20, 35} 16.50935
Log-likelihood -3.143516
Schwarz criterion 90.31342
rho 0.355342
S.D. dependent var 0.334226
S.E. of regression 0.323744
Within R-3quared 0.313524
F-value(F) 2.45e-12
Akaike criterion 43.23703
Hannan-Quinn 64.77675
Durb i n-Wat s on 1.059902
Figure 1. Model 1
growth rates. In our model we assume that using of the growth rate in percentage is the most reasonable approach.
Size of stockmarket within a country
As another indicator for the market situation of the country we include the size of the stockmarket. In fact it is discribed by the value of shares traded, both domestic and foreign, multiplied by their respective matching prices.
Money and quasi money growth
Money and quasi money comprise the sum of currency outside banks, demand deposits other than those of the central government, and the time savings, and foreign currency deposits of resident sectors other than the central government. This definition is frequently called M2. The change in the money supply is measured as the difference in end-of-year totals relative to the level of M2 in the preceding year.
Dummy Variables
Because of the reason we have four different countries with different data at the beggining we decided to use dummy variables to differenciate the countries. As we have 4 countries we have to use 3 dummy variables. But during our work we decided that the use of panel data is a more elegant way in our case and that because of this modification we can avoid manual introducing of dummies in the model. For this reason in our lattest model dummy variables explain not countries but years, because the differenciation of countries is already included through cross-sectional units in panel data tec-niques of Gretl.
4.2. Model specification
Following the literature and experts' recommendations and assuming our thoughts presented in the Literature review section, the basic specification of the model is:
Model 4: Fixed-effecta, using 43 observations Included 4 cros3-3ectional units Time-serie3 length = 12
Dependent variable: 1 AmountofMAdealsinnumbers Omitted due to exact collinearity : dt_14
coefficient std. error t-ratio
p-value
const 3. ,37904 0 , .159123 52 . .66 9. , 36e-029 ***
LNexch 2 -1. ,99466 0 , .513532 -3. .347 0. , 0007 ft* *
Maneyandquasimon~ -0. ,0216951 0 , .00736552 -2. . 345 0. , 0066 ft ft ft
Moneyandquasim- 1 -0. ,0213630 0 , .00764173 -2. .362 0. , 0030 ft**
Moneyandquasim~ 2 -0. ,0160080 0 , .00733679 -2. .167 0. , 0392 ft ft
Stockstradedto- 2 0. ,00434553 0 , .00157335 2. .762 0. , 0102 ft ft
GDPGrowthrate 2 -0. ,0439731 0 , .0175220 -2. .795 0. , 0094 ft ft ft
dt 3 -0. ,343959 0 , .162120 -2. .769 0. , 0100 ft ft
dt 4 -0. ,290113 0 , .159355 -1. .321 0. , 0793 ft
dt_5 -0. ,256555 0 , .165107 -1. .554 0. , 1319
dt 6 0. , 0379349 0 , .175220 0. .2163 0. , 3300
dt_7 0. ,347955 0 , .179230 2. . 399 0. , 0133 ft*
dt 3 0. ,522390 0 , .163611 3. .101 0. , 0045 ft ft ft
dt_9 0. ,112194 0 , .192517 0. .5323 0. ,5649
dt 10 0. ,275335 0 , .162165 1. .701 0. , 1004
dt 11 0. ,292100 0 , .170059 1. .713 0. ,0973 ft
dt 12 0. , 0301314 0 , .165300 0. . 3336 0. , 6326
dt 13 -0. ,139554 0 , .152231 -0. . 3164 0. , 3676
Mean dependent var 7.153632 5.D. dependent var 0.31342 7
Sum squared resid LSDV R-squared LSDV F (20, 27) Log-likelihood Schwarz criterion rho
1.071642 5.E. of regression 0.199225
0.965540 Within R-squared 0.914113
37.32539 P-value(F) 3.56e-15
23.13917 Akaike criterion -4.273331
35.01639 Hannan-Quinn 10.5713 9
0.392333 Durbin-Watson 1.146004
Joint test on named regressors -
Test statistic: F(17, 27) = 16.905
with p-value = F(F(17, 27) > 16.905) = 3.42207e-010
Te3t for differing group intercepts -
Null hypothesis: The groups have a common intercept Test statistic: F(3, 27) = 114.247
with p-value = F(F(3, 27) > 114.247) = 1.36595e-015
Distribution free Weld te3t for heteroskedasticity -
Hull hypothesis: the units have a common error variance Asymptotic test statistic: Chi-3quare(4) = 3.16413 with p-value = 0.530744
Figure2. Model 2
MA = p0 + ^Ex + P2GDPGr + P3Stock + P4M + u. (1)
As we have time-series observations for the same objects (the same countries) the data should be considered as pure panel data (each observations through time). So, we have 4 cross-sectional units (Russia, Brazil, China and India) with time-series length of 14. The total amount of observations equals 56.
5. EMPIRICAL RESULTS
The model was estimated through the panel data technique of fixed effect. For using it we had to
introduce a new variable (a) in the model, eliminating P0:
MA = P:Ex + P2GDPGr + P3Stock + P4M + a + u. (2)
The results of this specification are presented in the figure below (Figure 1). We discovered that all our variables apart from M (money and quasi money growth) are not statistically significant. Because of the low p-value for Ex that is 0.4749 and is much higher than the critical value of 0.05, we didn't find out the expected effect of exchange rates on M&A deals.
These results drove us to think more deeply about specification of the model and to modify our variables.
After some attempts of specifying the model we figured out that the most significant result could be obtained by including long-term effects of the factors (using lags). So, the best specification can be described as following:
MA = P:Ex_2 + P2GDPGr_2 + P3Stock_2 + + P4M + P5M_1 + P6M_2 + a + u. (3)
Emperical resultes are presented in the figure (Figure 2). The p-values for all variables (apart from time dummies) are lower than 0.05, so our variables are statistically significant. The R2 is 0.9655 what means that 96,55% of the dependent variable is described by the model, that is very high and indicates a high Goodness of fit. The joint significance of the model is also satisfying as the P-value for F-test is much lower than 0.05.
6. CONCLUSION
Following the results of our final model we can conclude that there is a significant relation between the exchange rates and the M&A market. We established this significant connection by introducing lags of mostly two years. We found this result surprising because we expected that the M&A and their analysts would react in a quicker way. For interpreting the coefficient of our model we have to take into account that our dependent variable is in logarithm. In the case of the exchange rate we can see that in our model we have a negative relation between the amount of M&A deals and the exchange rate. In our final model we interpret that if the exchange
rate increases by 1% the amount of M&A deals will decrease in the second following year by 1.99%.
Our results are opposite to those obtained by Hyan and Kim (2010), in whose model the coefficient for exchange rates appeared statistically insignificant. As well our results are in contrast with the paper of Baker, Foley and Wurgler (2009). The authors didn't find the existence of a cheap asset effect on FDI flows. But as we supposed in the beginning, dealing exactly with M&A and not with FDI in general allowed us to establish a correlation between costs of assets (expressed through exchange rates) and investment flows.
Despite the fact that the results obtained in this work do not agree on previous researches, the negative relation of exchange rates and M&A seems economically logical and fits with our initial expectations.
Nevertheless, our model has some limitations. One of them is that for our paper we used only data for BRICS countries. To extend the investigation it could be interesting to compare our results with estimations obtained for other groups of countries (e.g. developed, PIIGS (Portugal, Ireland, Italy, Spain), emerging etc.). Other limitation is the number of periods observed, because BRICS countries do not have a long history of established M&A and financial markets (e.g. Russia's market starts its existing only after the dissolution of the Soviet Union).
To extend the model the monthly data can be used, other countries can be included and the time period can be increased. The R-square of our model is rather high (99,55%), but maybe it can be increased by including some other variables. Other way to continue our study is to estimate M&A markets not in numbers but in value terms and then compare if the results are quite similar.
REFERENCES
1. Baker, M., Foley, C. F., & Wurgler, J. (2009). Multinationals as arbitrageurs: The effect of stock market valuations on foreign direct investment. Review of Financial Studies, 22 (1), 337-369.
2. Brooks, R., Edison, H., Kumar, M. S. & Sl0k, T. (2004). Exchange rates and capital flows. European Financial Management, 10 (3), 511-533.
3. Hyun, H. J., & Kim, H. H. (2010). The Determinants of Cross-border M&As: The Role of Institutions and Financial Development in the Gravity Model. The World Economy, 33 (2), 292-310.
4. Mileva, E. (2008). The impact of capital flows on domestic investment in transition economies. European Central Bank, Working Paper Series, no. 871.
5. Neto, P., Brandao, A., & Cerqueira, A. (2010). The macroeconomic determinants of cross-border mergers and acquisitions and greenfield investments. UP Journal of Business Strategy, 7 (1/2), 21.
6. Wong, W. K. (2008). Comparing the fit of the gravity model for different cross-border flows. Economics Letters, 99 (3), 474-477.
Annex 1. Data gathered for BRICS countries
№ Year Amount of M&A deals (in numbers) LN(excht)-LN(excht-1) Money and quasi money growth (annual%) Stocks traded, total value (% of GDP) GDP Growthrate (%) Country
1 1994 97 2.85447198 1102.383252 14.649049 5.334551702 Brazil
2 1995 153 0.322523208 44.30215492 10.04305444 4.416731354 Brazil
3 1996 191 0.091008102 31.03490423 13.65531198 2.207535524 Brazil
4 1997 233 0.070012703 17.24332067 24.61494984 3.39502864 Brazil
5 1998 387 0.073765566 12.01934813 20.01362137 0.338356177 Brazil
6 1999 353 0.446632023 18.1153445 26.50197649 0.469066589 Brazil
7 2000 530 0.008503352 19.70463724 14.34974741 4.112564911 Brazil
8 2001 408 0.250257913 14.35346436 11.52665705 1.657817967 Brazil
9 2002 258 0.217449333 9.861412337 7.751491745 3.05316092 Brazil
10 2003 212 0.052401564 20.45463125 12.62335851 1.140319046 Brazil
11 2004 270 -0.050774177 16.62980714 16.98484861 5.760880726 Brazil
12 2005 273 -0.183639091 18.46659893 19.16938816 3.202051527 Brazil
13 2006 377 -0.112517382 17.97593593 25.23172582 3.960502029 Brazil
14 2007 871 -0.110859158 18.67847251 46.19898042 6.072283693 Brazil
15 2008 940 -0.059947547 17.77519265 33.60385371 5.093767007 Brazil
16 2009 530 0.086489087 16.30237029 42.46300374 -0.12614741 Brazil
17 2010 712 -0.127986883 15.81598252 41.11292333 7.528797377 Brazil
18 2011 864 -0.050358285 18.50999909 31.55071272 3.910255352 Brazil
19 2012 836 0.154885724 15.90464201 33.79741521 1.915458618 Brazil
20 2013 629 0.09889422 8.912126126 29.99781629 3.015140514 Brazil
21 2014 566 0.087374692 13.53125026 26.65571032 0.103371356 Brazil
22 1994 106 0.402661806 31.50013453 12.12472219 13.07807061 China
23 1995 120 -0.031508027 29.4610222 10.59090248 10.99384345 China
24 1996 191 -0.004469296 25.2731568 35.74852876 9.924722663 China
25 1997 302 -0.002934036 20.72731167 38.72579973 9.226887728 China
26 1998 357 -0.001310699 14.90435007 27.73786276 7.853489523 China
27 1999 340 -8.55619E-05 14.66647771 18.80891678 7.618173474 China
28 2000 530 3.07025E-05 12.32478198 62.44152103 8.42928216 China
29 2001 570 -0.000173456 15.04241351 34.73708447 8.298374411 China
30 2002 1064 -1.33905E-05 13.14043628 23.13473555 9.090909091 China
31 2003 1704 9.56466E-06 19.23976666 23.51627896 10.01997337 China
32 2004 2400 -2.84929E-05 14.88692014 26.34206713 10.07564297 China
33 2005 1951 -0.010015696 16.7416524 17.29555186 11.35239142 China
34 2006 2212 -0.027325015 22.11611885 42.4572463 12.6882251 China
35 2007 2963 -0.046976934 16.73553458 178.9747162 14.19496167 China
36 2008 3408 -0.090590759 17.77810755 85.66670368 9.623377486 China
№ Year Amount of M&A deals (in numbers) LN(excht)-LN(excht-1) Money and quasi money growth (annual%) Stocks traded, total value (% of GDP) GDP Growthrate (%) Country
37 2009 3089 -0.017016135 28.42327787 154.7761808 9.233551095 China
38 2010 3721 -0.008991156 18.94831461 136.7253186 10.63170823 China
39 2011 4103 -0.046685321 17.32296979 89.07634935 9.484506202 China
40 2012 3810 -0.023350192 14.39165202 59.41175899 7.750297593 China
41 2013 3964 -0.018640391 13.58890221 81.09054129 7.68380997 China
42 2014 5122 -0.008481036 11.01193614 115.4951568 7.268460929 China
43 1999 423 0.042610199 17.14918048 0 8.845755561 India
44 2000 895 0.042875664 15.17170763 4.606709729 3.840991157 India
45 2001 721 0.048742035 14.32055069 30.74621329 4.823966264 India
46 2002 599 0.029729821 16.76116474 24.52523941 3.803975321 India
47 2003 723 -0.042594069 13.03361109 43.99294975 7.860381475 India
48 2004 790 -0.027571299 16.73233295 54.43146629 7.922936613 India
49 2005 1283 -0.027211253 15.5999039 55.60436663 9.284831507 India
50 2006 1524 0.027002514 21.63314112 68.67261762 9.263958898 India
51 2007 1570 -0.091424779 22.27150287 92.30519222 9.801360337 India
52 2008 1503 0.050843137 20.49520988 75.60304759 3.890957062 India
53 2009 1372 0.106728535 17.99583922 79.87247579 8.479786622 India
54 2010 1451 -0.056945668 17.80217706 63.27669484 10.25996299 India
55 2011 1116 0.020448604 16.13758934 35.1585227 6.63835345 India
56 2012 1169 0.135396197 11.04569666 33.63248973 5.081417925 India
57 2013 1022 0.092190172 14.83153 28.88478265 6.899217233 India
58 2014 1155 0.040659663 10.5873816 35.6698877 7.286253239 India
59 2001 398 0.036283219 35.84545974 9.198125911 5.091984231 Russia
60 2002 403 0.07207567 33.72158294 13.81221622 4.743669897 Russia
61 2003 501 -0.021163039 38.32511281 18.52522667 7.295854331 Russia
62 2004 406 -0.063150434 33.74554283 20.08296921 7.175949192 Russia
63 2005 477 -0.018540526 36.39268629 19.35632049 6.376187027 Russia
64 2006 699 -0.039427385 40.38872099 58.85758395 8.153431973 Russia
65 2007 999 -0.06104066 40.57945254 98.25464613 8.535080209 Russia
66 2008 1783 -0.028870404 14.3331788 69.54364405 5.247953532 Russia
67 2009 3357 0.244615565 17.31984985 41.74476562 -7.820885026 Russia
68 2010 3775 -0.04420237 24.588653 33.23799815 4.503725625 Russia
69 2011 3312 -0.032992774 20.86233565 29.0966645 4.264176566 Russia
70 2012 2610 0.04841322 12.07389426 16.87979921 3.405546804 Russia
71 2013 2096 0.031826567 15.65641834 11.32790537 1.340797614 Russia
72 2014 1958 0.186856129 15.45453814 8.59614579 0.640485765 Russia
Source: http://data.worldbank.org/.
