About as

At KMR clarity and breadth of vision through bright, challenging thinkers, and an uncommon affinity with our clients makes supporting long-term aims second nature to us. We’re strategists, not box-tickers. 

We care. We’re interested. 

We bring ideas to you. 

Our growth in our business vindicates our philosophy of quality first, and it is a proof point for the efficiency of the in-house systems that we’ve developed over the past four decades to ensure quality compliance work, on time, every time. If you are looking to work with us, you’ll find in us an ally as much as an accountant.

Estimation methodology

All explanatory variables used in our empirical model are not strictly exogenous. If this is the case, straightaway panel estimation will yield biased results. Therefore, a test for endogeneity should be applied. If the null hypothesis of exogeneity is rejected, then Arellano– Bover (1995)–Blundell Bond (1998) system GMM estimator will be employed. GMM allows for controlling the joint endogeneity of explanatory variables through the use of internal instruments (see Enders, 2004 and Baltagi, 2005). The tricky issue in GMM methodology is to select valid instruments/moments. No rule of thumb exists in instruments' selection. For this purpose, Murray (2006) discussed various tricks. Two diagnostic tests are there for the validity of instruments. Firstly, we have Hansen test for over-identifying restrictions with the null hypothesis that the instruments are not correlated with the residuals. Secondly, we have Arellano–Bond test for second-order correlation in the first differenced residuals. The advantage of using GMM over the other instrumental variable (IV) methods is that a G MM estimator is more efficient if heteroskedasticity is present. Similarly, a GMM estimator is not worse than an IV estimator if heteroskedasticity is not present. In this study, the lagged values of independent variables have been used as instruments. We also perform a robustness check by the inclusion of additional control variables in our model.

We apply the mentioned estimation technique in this study on aggregated and disaggregated data (by decomposing global sample into various groups like Asia, Europe, and Africa) for various periods (1996-2017, 2013-2017, and 2008-2012). The intention for this disaggregation is to watch the role of corruption in growth and public debt relationship for BRI countries closely. Similarly, the purpose of dividing period is to see the impact of BRI initiative on the relationship.

 

To see the impact of BRI on growth, we use fixed and random effect models by including period dummy on the aggregate data (global sample). For this purpose, we will divide data into two parts, i.e., 2008-2012 and 2013-2017, by assigning 0 to the former and 1 for the later. Comparative to the random and fixed effect models that are not restricted, the pooled model is restricted and assumes that countries are homogeneous. When it is necessary to control for omitted variables that are constant over time but differ between countries, the fixed-effects model is desirable. Since the fixed effect considers heterogeneity and individual country effects, therefore, it gives better estimates than the pooled model.

On the other hand, no individual country effects are assumed in the random effect model. Hausman (1978) test is employed to choose between fixed- and random-effect models. Hausman test specifies whether the explanatory variables are correlated with specific effects or not. Hausman test makes sure the selection of the model with consistent results. Random effects are not correlated with the explanatory variables is the central assumption in random effect estimation. The fixed effect model is feasible if the p-value is significant, i.e., < 5%. On the other hand, if it is greater than 5%, then the most appropriate model is the random-effects model.

Furthermore, we employ quantile regression (QR) to examine influence of public debt on income distribution in BRI countries. For comparison purpose, we also employ the fixed effect model. QR estimation methodology enables the estimation at different intervals of condition distribution.

Model specification

We analyze the debt-growth relationship in the neoclassical growth framework. There is always an issue of heterogeneity in cross-country panel analysis; therefore, convergence issue arises. Economies with higher values of real income tend to grow faster than the economies with a higher starting value of real per capita income (Barro and Sala-i-Martin, 2004). However, we use government debt-to-GDP ratio. Rather than the initial per capita income, other aspects also can explain the convergence phenomena.

 

Our study is designed to estimate the linearities and nonlinearities in the debt-growth relationship in the context

of BRI countries. In this paper, we examine the impact of various indicators of debt burden on the economic growth

of BRI countries and corruption indices as the transmission channels. Our models take the following forms;

                    

Where grow is the GDP growth, Xitj is a set of control variables, DEBT is public debt, and DS is debt servicing. Subscripts i and t represent panel and time dimension, while η and  denote time-specific and country-specific effects.

Data and estimation methodology 1- Data

The basis concern of our study is to explore the relationship between public debt, economic growth, and income inequality with the transmission channel of corruption. We use annual data from 1996-2017 relating to BRI countries. The BRI is a newly established economic block and consist of countries from Asia, Europe, Africa, Oceania, and Latin America (see Appendix for list of countries). More than seventy countries are included in BRI. Due to data limitation, our sample restrict to only sixty countries.

Our main variables of interest are public debt and GDP growth. The corruption indices are sourced from Kaufmann et al. (2013) and Transparency International (TI). We include the two kinds of corruption indices to ensure robustness of results. Kaufmann et al. (2013) corruption index ranges from -2.5 (totally corrupt) to 2.5 (not corrupt). The value of Corruption Perceptions Index (CPI) of TI ranges from 0 (totally corrupt) to 10 (not corrupt). The scales of the two indices are reversed for estimation and maintain consistency between the two indices. We reverse the scale for Kaufmann corruption index with 5 for totally corrupt and 0 for not corrupt country. The TI is rescaled with 0 stands for not corrupt and ten totally corrupt.

Moreover, we include shadow economy as one of the controlled variables in our study. The reason for the inclusion of shadow economy is that corrupt countries are having large shadow economies. The shadow economies affect the level of economic growth and public debt. For shadow economies, we rely on the data of Schneider et al. (2010). Schneider et al. (2010) argue that shadow economies become large due to several elements. The elements consist of increasing taxation and higher regulation with lower institutional quality. They use the Multiple Indicator Multiple Cause (MIMIC) approach to estimate the shadow economy because the shadow economy cannot be measured directly. The data run from 1999-2007; therefore, the missing data is interpolated till 2017.

 

We include foreign direct investment (FDI) as one of our control variables. A plethora of literature shows FDI and economic growth relationship. Several macro-based articles on both developed and developing countries indicate a positive effect of FDI inflows (Olofsdotter1998; Reisen and Soto 2001). However, other studies report an adverse effect of FDI on economic growth (Mencinger2003; Carkovic& Levine 2005; Johnson 2006; Türkcan, Duman, and Yetkiner2008; Herzer 2012) or an inconclusive effect (De Mello 1999).

Moreover, infrastructure also plays an essential role in economic performance. We include global infrastructure index in this study as our control variable. The construction of the global infrastructure index is explained in Donaubauer et al. (2015). The index is based on transport, energy, ICT (internet and communication technologies), and financial indicators. The data is available from 1990–2010, the rest of it is interpolated. Furthermore, secondary school enrolment and inflation are used as a proxy for human capital and macroeconomic stability. Details of all variables along with their description statistics are given in Table 1.

Table 1. Data source and summary statistics

Variable	Notation	Obs.	Mean	SD	Min	Max	Comment
GDP growth (annual %)	grow	1,652	4.41	5.12	-37.14	54.15	WDI (2018)
Public Debt% of GDP	GD	955	5.3	0.4	2.3	6.5	WDI (2018)
Corruption Index Kaufmann et al. (ranges
from approximately 0 (no corruption) to 5	KCI	1,680	2.57	1.11	0.98	5	Kaufmann et al. (2013)
(high corruption)
Corruption Perceptions Index TI. (ranges	TI	1,680	2.98	1.43	0.00	8.8	Transparency International,
from 0 (not corrupt) to 10 (totally corrupt)							2013
Shadow economy% of GDP	shadow	1,510	3.4	0.5	0.7	4.2	Schneider et al. (2010)
Global infrastructure index	GINFRA	1,652	0.8	0.4	-2.3	1.7	Donaubaueret al. (2015)
Total debt service (% of GNI)	DS	1,064	1.5	0.8	2.3	3.6	WDI (2018)
Foreign direct investment, net inflows (%	FDI	1,680	3.7	0.2	2.3	4.5	WDI (2018)
of GDP)
Inflation	INF	1,680	7.8	0.3	2.3	9.8	WDI (2018)
GINI index	GINI	1,056	3.56	0.21	2.79	4.17	WDI(2018)
GDP per capita (constant 2010 US$)	GDPPC	1,652	8.51	1.30	5.1	11.19	WDI (2018)
Secondary school enrolment (gross %)	HC	1,653	4.89	0.28	-2.30	5.37	WDI (2018)
Gross capital formation (% of GDP)	GCF	1,652	4.92	0.21	-2.30	5.44	WDI (2018)
General government final consumption	GE	1,652	3.35	0.28	-2.30	4.49	WDI (2018)
expenditure (% of GDP)
Urban population (% of total)	URBAN	1655	1.64	0.20	1.07	1.95	WDI (2018)
Exports plus imports divided by GDP	OPEN	1,644	1.77	0.51	-0.77	2.33	WDI (2018
School enrollment, secondary (% gross)	HC	1,653	4.9	0.3	2.3	5.4	WDI (2018)

Note: All variables have been converted into a logarithmic form for the empirical estimation except the corruption indices.

Before proceeding further, we need to identify the order of integration of variables in order to avoid a possible problem of spurious regression. Table 3(see Appendix) reports the results of the unit root. For robustness check, we report four different kinds of tests’ results. All the results show that our variables are stationary at level.