A dynamic process of forming complex structured economic entities at the sectoral, interindustry, and interregional levels based on the integration policy takes place in the modern economy. A structural change happens in mergers and acquisitions transactions; they involve an increasing number of participants; the scale of cross-border transactions is expanding, the state becomes an active participant. At the same time, the nature and complexity of integration processes require qualitative tools for a comprehensive analysis of the integration activity of Russian companies. The aim of the research presented was the development of an econometric approach to the analysis of integration policy in the Russian economy. Multidimensional statistical and econometric methods for analysing the dependencies, reducing the dimension and classification, as well as the system of econometric equations were used as a research tool. The econometric approach is based on a recursive system of simultaneous equations describing the integration activity of Russian companies and the impact of various macroeconomic indicators on it. The analysis of the model obtained proves a unidirectional change in the volume of foreign investments and the value of mergers and acquisitions transactions, as well as the inverse relationship between the volatility of the stock market and the value of integration transactions. The results of the research are of practical importance, since they can be used in order to identify favorable conditions for doing business, improve the investment climate, as well as to increase the validity of decision-making on the socioeconomic development of Russian territories.
The role of integration policy in the system of economic relations of any state has significantly increased in recent decades. The scale and level of economic in? The work is performed under the grant of the President of the Russian Federation for the state support of young Russian scientists PhDs (МК–5339.2016.6). tegration are largely macroeconomic indicators of the effective functioning of the national economy and its institutions. Integration processes under modern Russian conditions restore the structural integrity of the national economy, equalise the spatial characteristics of the country’s industrial potential, activate the innovative business activity, and increase the competitiveness of domestic products that is the key to the reindustrialisation of the domestic economy.
The processes of integration policy are the objects of close attention and numerous studies. However, the issues of economic content and quantitative measurement of the integration activity in the Russian economy remain insufficiently developed. Moreover, it is impossible to reindustrialise the Russian economy successfully, as well as to create an effective economic system with a fundamentally new nature of corporate relationships without statistical consideration of the factors that determine the integration activity of Russian companies. In this regard, there is a need to model the processes of mergers and acquisitions of Russian companies based on modern econometric approaches. 2
Econometric analysis of mergers and acquisitions
processes in the Russian economy
Integration processes in business aimed at its scale and market share increasing
become an important factor in raising competitiveness under the conditions of
Russia’s imposing economic sanctions and increasing international isolation. The
research of the mergers and acquisitions (M&A) processes and the integration
activity in various states with an advanced institutional environment is largely
based on econometric methods and models [
Thus, Choi S. H. and Jeon B. N. studied the dynamic impact of
macroeconomic factors on the integration activity in the USA economy for the period
1980–2004 in their work [
In the 1990’s, the importance of the macroeconomic environment influence
on the location of international production in the host country (M&A export
transactions) against the intensification of globalisation processes was widely
discussed in foreign scientific literature. Thus, Hughes H., Uddin M., and
Nottingham M. in the article [
A relatively short time series of domestic databases on M&A transactions,
which can give a very diffuse assessment of the Russian market of external
mergers and acquisitions, is one of the serious limitations for the use of this kind of
model in Russia. Insufficient development of tools of the Russian stock
market for transaction payment may be another restriction apart from the sample
size [
Bhattacharjee A. and Higson S. in their work [
Thus, a large number of foreign authors devoted their works to the dynamics
of macroeconomic indicators and its impact on the intensity of mergers and
acquisitions processes. At the same time, this research direction is relevant also for
the Russian economy. Although many models of this direction have not yet been
realised under Russian conditions, nevertheless it has recently become possible
to analyse the relationship between Russian corporate integration processes and
a number of macro variables [
Ignatishin Yu. V. was the first who used econometric methods for the study
of the market for corporate control [
Musatova M. M. proposed a regression model of relationship between macro
variables and the dynamics of M&A transactions over 2001–2004 [
At the same time, an assessment of the close relationship between individual variables and the creation of regression equations is not enough in order to describe the mechanism of the functioning of the market for mergers and acquisitions. To characterise the true impact of individual characteristics on the change in the system of the resulting indicators of the M&A market it is not reasonable to consider the multiple regression equation taken separately. In this case, these processes can be described with systems of interrelated (simultaneous) equations.
For a complex analysis of the market of mergers and acquisitions it seems necessary to use 3 resulting variables with a causal and consecutive relationship based on information collected monthly from January 2003 to December 2015 (156 observations): y1 the cost of conflict assets; y2 the number of integration transactions; y3 the cost of integration transactions.
Various factors that determine the specificity and efficiency of the use of integration strategies for growth and development by business entities influence the integration processes. The internal prerequisites for mergers and acquisitions may include: the achievement of certain financial indicators that are necessary for the further development of the company and its transition to a qualitatively new level; the correction of a company’s financial position, when integration with other market participants will allow to solve a number of unsolvable problems.
The ongoing macroeconomic processes can be attributed to the external
factor that influences the development of mergers and acquisitions processes. The
impact of the stock market and its capitalisation on the M&A market is that
one of the main integration schemes is based on the acquisition of shares of the
target company [
According to the experts of the committee on corporate finance and financial management of the audit and consulting firm FAC, investments are one of the indicators of the mergers and acquisitions market movement. Currently, Russia lags behind most post-Soviet countries in terms of the level of investment for infrastructure development that is reflected in the fact that the share of foreign buyers in the total value of mergers and acquisitions transactions has been recently decreasing.
According to the research of the CMS consulting company and the Mergermarket analytical agency, which was based on the survey of top managers among 32 large companies and 100 participants of the Russian market of mergers and acquisitions in 2015, the institutional reforms in the economy and, particularly, the state privatisation program have impact on the market of mergers and acquisitions in the Russian Federation. The government intends to sell minority block of shares in the leading state-owned enterprises in order to reduce the budget deficit, increase the investment attractiveness of assets, and attract serious private investors in privatised companies.
According to the press service of the government, in February 2017, the head of the government approved the forecast privatisation plan for 2017–2019. The budget with revenues from privatisation without taking into account the sale of shares in the largest companies should reach 5.6 billion rubles annually. It is planned to privatise 477 JSCs, 298 FSUEs, the Russian Federation’s stake in 10 PLCs, as well as 1041 property of the state treasury for the period 2017– 2019 in total. In particular, it is envisaged that Russia will cease to participate in Novorossiysk Commercial Sea Port PJSC, United Grain Company, Prioksky Non-Ferrous Metals Plant, and Crystal PA.
When characterising the Russian market, a special attention should be paid
to crimes and offenses in the sphere of economic activity. Clever abuse of
procedural rights during corporate conflicts entails the difficulty and frequently the
complete paralysed activities of economic entities of the Russian market that
causes significant material losses [
Thus, the factors that directly influence the intensity of mergers and
acquisitions processes (see Fig. 1) for the period from January 2003 to December 2015
were divided into 6 functional blocks: macroeconomic indicators (11 variables);
corporate finance (12 variables); the Russian stock market (12 variables);
investments (8 variables); institutional changes in the economy (7 variables); crimes
and offenses in the economic sphere (4 variables). At the same time, it should
be noted that in reality the causal relationship between the integration
activities and selected regressors is multiple-valued [
Further, the Granger causality test was used to test these assumptions for
causal dependency; this test includes the application of the Fisher test which is
used to check whether the lag information about the variable x has a statistically
significant effect in explaining yt at accounting explanatory variables and lagged
values of y [
m yt = 0 + X i=1
m iyt i + X i=1 ixt i + "i:
If lagged x in the presence of lagged y does not make a statistically significant contribution to the explanation of yt, then it is considered that “x does not Granger cause y”. Alike, if lagged y does not contribute a statistically significant input in explaining xt in the presence of lagged x:
k xt = 0 + X j=1
k jxt j + X j=1 jyt j + "j; then it is considered that “y does not Granger cause x”.
The hypotheses presented in Table 1 were put forward to clarify the causal relationship between the resulting factors.
As a result of the analysis, it was found that the value of conflict assets is a Granger cause for the number of mergers and acquisitions transactions and the monetary value of the M&A market, but the number of integration transactions is a Granger cause for the monetary value of the M&A market. All abovementioned shows that the model of the integration activity in the Russian Federation can be represented in the form of a recursive system of simultaneous equations.
Based on the method of eliminating quasinemeasurable variables at a critical value of the coefficient of variation = 0:1, the following 8 characteristics with no significant information were recognised as quasi-unchangeable and excluded from the set of potential explanatory variables: the consumer price index; producer price index of manufactured goods; index of industrial production; the producer price index of industrial goods; official exchange rate of the Ruble to US dollars; the number of shares traded on the Russian stock market; the number of issuers traded on the Russian stock market; the number of crimes committed in the economic sphere.
Functional and weakly bound regressors were identified with respect to endogenous variables based on the method of analysing the matrix of correlation coefficients (Table 2). 2.1
The creation of the econometric model of the indicator y1 the value of the conflict assets With the help of cross-correlation function, lagged variables were created based on the exogenous variables. Lag value for variable xj was defined from r(y1t; xjt ) = max(r(y1t; xjt )), varying from 0 to 12. For instance, the volume GDP for lag x1 was 8 months. Similarly, lags were defined for the remaining exogenous variables xj analysed. (1) (2)
At the same time, a repeated analysis of the matrix of paired correlation
coefficients of the 15 remaining characteristics showed the presence of
multicollinearity. In order to preserve the number of endogenous variables for further
economic interpretation, the transition to an orthogonal coordinate system was
used [
The implementation of the principal component method with the subsequent orthogonal rotation resulted in the five generalised factors that explain 83.04% of the total variance. The regression equation for y1 the value of conflict assets, created based on the individual values of the generalised factors f1 f5, has the form:
All regression coefficients in the equation are significant at the level = 0:05. The parameters of the equation obtained show its statistical adequacy: Fobserv > Fcrit with = 0:05, found under the table of the F -distribution. The standard error was 0.28.
Based on the factor load matrix that characterises the close relationship between the characteristics and the main components, as well as the matrix of eigenvectors with respect to the original variables, the regression equation takes the form: y^1;t = 0:12 + 0:23y1;t 1 + 0:07x1;t 8 where R2 = 0:8415, Fobserv = 53:03, s^ = 0:39.
To investigate the presence of autocorrelation in the remainder, an
asymptotic criterion of the Breusch–Godfrey serial correlation was used, which is based
on the idea that if there is a correlation between neighboring observations, then
it is natural to expect that in equation (3) the coefficient turns out to be
significantly different from zero [
et = t + et 1; (3) where et is the remainder of the regression equation.
According to available data, the coefficient = 0:019. It is not significantly different from 0, therefore there is no autocorrelation in the remainder. 2.2
The creation of the econometric model of y2
integration transactions.
the number of
The analysis carried out made it possible to reveal that since the instant in time
t = 69 (September 2008) a structural change takes place in the character of the
dynamics of the indicator under study. This instant in time is characterised by
the beginning of the financial and economic crisis in Russia. Buyers and sellers
have different reasons for mergers and acquisitions during the periods of economic
growth from reasons for integration during the financial and economic crisis
[
In econometrics, several formalised tests have been developed that allow one to determine the presence of a structural shift in the available data. In this paper, the Chow test was applied, which showed that a structural shift was observed in September 2008. Therefore, the parent population was divided into two parts in terms of improving the model quality relative to the instant in time t = 69.
Further, with the use of the cross-correlation function, the lag variables were created based on the exogenous variables. Thus, the lag for x7 the deficit/proficit of the consolidated federal budget is 7 months. The lags for other exogenous variables xj were created similarly.
The analysis of the matrix of paired correlation coefficients showed a high
multicollinearity in the independent variables. The approach to maximise the
predictive power of regression models [
To use the entire set of observations in the quantitative volume model of the mergers and acquisitions market, a dummy variable ut was used that takes the values 1 for all t < t and the value 0 for t t , i. e.
u =
The discrete nature of the dependent variable affords ground to assume that linear models that connect the number of mergers and acquisitions transactions with the levels of the factors accompanying them will not be entirely adequate to actual data because the calculated values y^2t can take both integer and fractional values. Count data models, in particular the Poisson regression model are more acceptable in such situations:
Yi = e 00xt+"t ; i. e. it is assumed that the number of events yt is distributed according to Poisson’s law with the parameter t = e xt .
The maximum likelihood method was chosen to create Poisson multiple regression (calculations were carried out using the Matrixer package). As a result, the following model count data was obtained: ln y^2;t = 2:03 + 0:11 ln y2;t 1 + 0:05 y^1;t 1 (4:02) (6:02) (4:47) 0:08 x16;t + 0:003x31;t + 0:12 x54;t + 0:01 ut: ( 8:05) (5:27) (4:01) (3:53)
The likelihood ratio criterion was used to test the hypothesis of the significance of Poisson regression. Since o2bserv = 62:37 > c2rit(0:05; = 1) = 3:84, then the created Poisson regression equation is significant as a whole.
According to the available data, the pseudo-coefficient of determination Rp2seudo = 0:8423, which shows that 84.23% of the variation in y2 is due to the factors included in the model. The Akaike information criterion, which takes into account the requirement of the increased model accuracy and the reduced number of model parameters, was AIC = 8:03.
The asymptotic criterion of Breusch–Godfrey serial correlation showed the absence of autocorrelation in the remainder. To check whether the remainders of the created model count data belong to the Poisson distribution, the following property was used: the variance of a random variable distributed according to a Poisson law is equal to its mathematical expectation:
M [y=xt] = D[y=xt] = : (4)
At that time if the values M ("t) and D("t) are equivalent, then this may serve as an argument in favor of the hypothesis of a Poisson distribution of remainder; the sharp difference in these characteristics, on the contrary, testifies against the hypothesis. Since M ("t) 2 and D("t) 1:43 then the hypothesis of the Poisson distribution of the remainder is accepted that testifies the model adequacy.
The calculation of the average marginal effects (Table 3) showed that the value of the conflict assets and the number of mergers and acquisitions transactions related to the previous instant in time has the maximum impact on y2. Moreover, the number of criminal cases related to the crimes committed during illegal seizures of property complexes of legal entities, as well as property and non-property rights of enterprises (raiding) also has a direct impact on y2.
The calculation of average marginal effects also showed a unidirectional change in the number of integration transactions and overdue creditor indebtedness, as well as overdue indebtedness for loans and borrowings among large and medium-sized enter-prises. This can be explained by the fact that the management of companies that have fallen into a crisis situation has two options: bankruptcy or business sale, and in most cases they prefer the second one.
At the same time, it is interesting that the nature of the dependence of the number of mergers and acquisitions transactions on the share of unprofitable enterprises is the opposite. This may be because a very popular way of seizure and acquisitions via the bankruptcy procedure was not used so often in the period under study. This results from the fact that a potential bankrupt company ceases to function in a normal mode and attention to it weakens.
The creation of the econometric model of y3 index number the cost of integration transactions.
The analysis revealed that, starting from the instant in time t = 65 (May 2010), a structural change occurs in the character of the dynamics of the indicator under study resulting in a change in the trend that describes this dynamics. This instant in time is characterised by the starting of changes in the global general economic situation and global factors.
The Chow test was used to test the hypothesis and it showed that it is advisable to divide the initial population into two parts for the further analysis from the point of view of the improved quality of the model relative to the instant in time t = 65 (January 2003–May 2008 and June 2008–December 2015).
As in the case of y1 and y2, the analysis of the matrix of paired correlation coefficients showed the presence of multicollinearity between independent variables. The use of the method to maximise the predictive power of the regression models allowed us to identify that the reduced set of indicators for y3 contains 10 endogenous variables: x1, x15, x16, x18, x24, x26, x31, x36, x37, x54.
A dummy variable zt, which takes the values 1 for all t < t and the values 0 for t t , was included to use the entire set of observations in the model of the monetary value for the mergers and acquisitions market, i. e. z =
The equation created indicates that the following indicators such as the proportion of unprofitable organisations in the total number of organisations in the industry, the volatility of the stock market, the volume of investments from foreign investors, and the number of criminal cases related to raiding influence y3 the monetary value for the mergers and acquisitions market. At the same time, the dummy variable zt, which characterises the structural instability of the resulting indicator y3t, is the most direct influence.
The asymptotic criterion of the Breusch–Godfrey serial correlation showed the absence of autocorrelation in the remainder. It can be seen from the histogram of the remainders that the remainders obey the normal distribution law.
Thus, the model of the Russian market of mergers and acquisitions can be represented in the form of a system of equations: 8>y^1;t = 0:12 + 0:23y1;t 1 + 0:07x1;t 8
The resulting system of interdependent econometric equations expresses their new content through their structural form the interrelationship between the phenomena; this content is characterised by the mutual influence of dependent and independent variables on each other, as well as allows for a deeper study of the causes of the relationship underlying the variation of the resulting variables. 3
Conclusion Each of the equations of the system obtained includes the number of criminal cases related to unfriendly and illegal acquisitions. These processes have an objective basis. Tendencies to increase the size of the money supply, which is the accumulation of both individuals and organisations, are characteristic for the modern Russian economy. The free money supply overhangs the mark, seeking an effective application for itself including via the new businesses acquisition that are often associated with criminal risks.
The application of the results of the calculations performed for the period from January 2003 to December 2015 allows one to see the opposite relationship between the volatility of the stock market and the value of the market for mergers and acquisitions. Thus, the increased uncertainty in the capital market reduces incentives for Russian holdings to implement investment projects, including integration projects. At the same time, the increased uncertainty in the stock market increases the value of conflict assets.