New Trends Between Governance and Growth
December 14, 2015
Using data from the World Bank’s Governance Indicators (WGI) and the World Bank’s Development Indicators (WDI), Kristijan Gjorgjevik, a PhD student in the economics program at Penn and Economics and Sustainability summer intern at the World Bank, illustrates some interesting new trends between Governance and Growth.
Author: Kristijan Gjorgjevik, G’17
What is Governance?
Governance refers to all processes of governing, whether undertaken by a government, market or network, whether over a family, tribe, formal or informal organization or territory and whether through laws, norms, power or language.
Kaufmann and Kraay (2002) define governance as “the process by which governments are selected, monitored and replaced; the capacity of the government to effectively formulate and implement sound policies; and the respect of the citizens and the state for the institutions that govern economic and social interactions among them.
For the first part of their definition of governance: the process by which those in authority are selected and replaced, Kaufmann and Kraay (2002) define two governance clusters: voice and (external) accountability and political stability. The former measures the extent to which citizens of a country are able to participate in the selection of governments, while the latter measures perceptions of the likelihood that the government in power will be destabilized or overthrown by possibly unconstitutional or violent means, including terrorism.
For the second part of their definition of governance: the government’s ability to formulate and implement sound policies, Kaufmann and Kraay (2002) define two more governance clusters: government effectiveness and regulatory quality. The former measures the quality of inputs required for the government to be able to produce and implement good policies and deliver public goods, while the latter is more focused on the policies themselves (This seems normative: Kaufmann and Kraay (2002) discuss market-unfriendly policies such as price controls or inadequate bank supervision, as well as perceptions of the burdens imposed by excessive regulation).
For the third part of their definition of governance: the respect of citizens and the state for the institutions that govern their interactions, Kaufmann and Kraay (2002) define two final governance clusters: rule of law and control of corruption. The former measures a society’s success in developing an environment in which fair and predictable rules form the basis for economic and social interactions, while the latter measures perceptions of corruption.
This set of six composite governance indicators are complied into a database called the Worldwide Governance Indicators (WGI). These indicators are inherently subjective. Kaufmann and Kraay (2002) argue that this is useful for two reasons. First, objective data are almost by definition extremely difficult to obtain. Second, perceptions of governance may often be as important as objective differences in institutions across countries. Beck et al. (2001) provide a compilation of objective indicators of political systems across countries, in the Database of Political Institutions (DPI).
Kaufmann and Kraay (2002) implicitly assume that within each of these six cluster, the indicators measure a similar underlying basic concept of governance. They use an extension of the standard unobserved components model to do so (see Goldberger (1972)).
What are the trends in the data?
Trend 1: Governance Indicators are positively correlated with GDP Per Capita and negatively correlated with GDP Growth and Volatility
The first trend suggests that nations that are governed better have a higher GDP per capita, but lower GDP growth. They also have lower GDP volatility. While this trend, in and of itself, says nothing about causation (i.e. whether governance is causing these economic phenomena, or vice-versa, or both), we could come up with some plausible stories about causation in either direction.
In the governance to GDP direction: 1) A primitive, extractive country could be intricately tied to its natural resources - small changes in commodity prices would have huge effects on GDP; 2) A primitive country could experience frequent changes in ruling party - different leaders may favor different industries, leading to spending surges followed by austerity measures; 3) A country with primitive infrastructure may experience frequent congestion, and incidents with its infrastructure that translate to fluctuations in productivity and output; 4) A country with bad governance may only be able to attract riskier private investments, given the desire of attaining a certain average rate of return. In the GDP to governance direction, the stories are less interesting. Causation in this direction is likely an effect of the subjective nature of the governance indicators; countries with higher GDP per capita and more stable growth could be more favorable ranked on the intrinsically subjective measures of governance.
In Figure 1, I plot the relationships between GDP Per capita and GDP Volatility for the six Governance Indicators:
Fact 2: Governance is highly correlated with future measures of GDP and FDI
In a first attempt at answering the causation question, I calculate in which set of years the GDP and Governance measures are highest correlated. In other words, it may be the case that Governance in 2007 has the highest correlation with GDP in 2005, or 2009. I find what difference in years, across the board, leads to the highest total correlation. The hope is that this provides some measure of which comes first – governance or GDP.
The plots below offer suggestive evidence that governance affects GDP growth and volatility, and FDI inflow. This is due to the high density of histogram barsto the left of 0, which indicates that governance indicators from the past have the highest correlations with GDP measures today.
Notice, as expected, that there is a very high density with the GDP per capita measure at 0. This is likely due to the subjective nature of the governance indicators; if times are good, individuals are likely to rate a country as having better “governance”. Indeed, more cross correlations are statistically significant with the GDP per capita measure (highest count is close to 80); this is consistent with reduced-form regression results in previous papers that show that GDP per capita is the most significant determinant of governance indicators (and risk ratings).
Fact 3: Voice and Accountability and Government Effectiveness are the only indicators that have a negative and statistically significant impact on GDP Volatility
Finally, I dissect the previous figure to show what time difference generates the largest negative correlation of the Government Effectiveness (GE) measure with respect to GDP Volatility. The takeaway message here is: GE is very noticeably negatively correlated with GDP Volatility (and growth) in future years. What does this mean, from a public policy perspective? It suggests that “the quality of public services, the quality of civil service and the degree of its independence from political pressures” is important; in particular, this suggests that governments that are less corrupt and offer more public goods engage in policies that generate lower volatility in economic growth. This is broadly consistent with some literature that finds that corruption increases growth rates in countries - it may do so, but at the cost of larger growth volatility.
I find a similar result with the Voice and Accountability (VA) measure. The public policy implication of this, is that “the extent to which citizens of a country are able to participate in the selection of their governments” is important; in particular, this suggests that governments that are more accountable to their people engage in policies that generate lower volatility in economic growth.
The next steps are to develop a model where some elements of governance (say, VA) are exogenous - this will capture underlying differences in institutions such as the political system and voting rights, which are hard to change by incumbent ruling powers in the short-run; and other elements of governance (say, GE or another proxy for corruption) are endogenous - this will allow for reverse-causality in the growth to governance direction. Hopefully, more to come soon!
Additional Blog Posts
Student Blog Disclaimer
The views expressed on the Student Blog are the author’s opinions and don’t necessarily represent the Penn Wharton Public Policy Initiative’s strategies, recommendations, or opinions.