Has China’s Investments Succeeded in Changing Country Governance Abroad?Rio RinaldiBlockedUnblockFollowFollowingJan 23Photo by chuttersnap on UnsplashChina’s foreign aid program now rivals some of the most industrialized nations within the OECD, and with their unorthodox approach of state-to-state lending, as well as very lax requirements when it comes to loans, many articles have pointed their finger at China for changing country governance, accusing them of making countries autocratic.
Is that really the case?This project analyzes China’s aid and investment projects from 2000–2014, and any impact it has had on changing the nature of government regimes, whether it means they became more democratic or autocratic.
I obtained the China aid and investment data from Aiddata¹, a huge dataset from the College of William and Mary detailing Chinese aid and investment projects all around the world from 2000–2014.
It has been extensively cited in many journalistic pieces given that this data is not exactly promoted and published by the Chinese government themselves and is a godsend for people interested in understanding China.
To look at government regimes I used the Polity IV series², a dataset from the Center for Systemic Peace that measures a range of regime characteristics such as democratic and autocratic rating, the participation of the public in choosing their executives, and a few others.
To start the analysis, I first combined both datasets, then cleaned and standardized them.
I then conducted exploratory data analysis.
From there, I decided to segment the number of cases into the year 2000 (prior) and 2014 (after) to see whether significant changes occurred.
I then conducted a few hypothesis tests, before finally concluding the findings in this analysis.
So let’s get to it!1 | Exploratory Data AnalysisWe’ll have to be a bit parsimonious here as the China aid dataset has over 70 columns, and although the Polity IV series was comparably modest, there is still a ton to explore.
I’ll share some visualizations that I thought were interesting and useful.
It’s probably no secret here, but China does primarily invest in energy and transportation infrastructure.
There is a clear upward trend, and the aftermath of the 2008 financial crisis seems a major turning point.
China managed to avoid the worst of it.
There is a slight upward trend towards governments that allow for genuine competition, but the trend is very modest.
The trends here are tougher to see.
It seems those repressed and suppressed are getting smaller, but the proportion of factional voting (e.
g.
, racial, ethnic) is slightly increasing.
Post-2008 seems to be when China’s reach became global and investment in far-flung countries (Africa & Americas) was consistent.
2 | Hypothesis TestingThe assumption that the polity data from 2000 and 2014 are independent is clearly a weak assumption, but it is one we will use for this analysis.
The good thing is that the distribution for the democracy ratings look decently normal, and given our sample size of 117 we can assume that the central limit theorem can do its job.
A decent normal distribution for both years.
Performing a two-sample t-test in R, we get the following results:Welch Two Sample t-testdata: democ by yeart = -2.
0543, df = 215.
59, p-value = 0.
02058alternative hypothesis: true difference in means is less than 095 percent confidence interval: -Inf -0.
191256sample estimates:mean in group 2000 mean in group 2014 4.
103774 5.
080357It seems that on an aggregate level, the countries that China gave aid to are significantly more democratic than in the past.
Let’s now segment the countries into the top 25th percentile of aid recipients and again conduct the t-test.
We now have a sample size of 33 countries.
Even better looking!Welch Two Sample t-testdata: democ by yeart = -0.
7731, df = 63.
513, p-value = 0.
2212alternative hypothesis: true difference in means is less than 095 percent confidence interval: -Inf 0.
7024852sample estimates:mean in group 2000 mean in group 2014 3.
575758 4.
181818The mean democracy rating for 2014 did increase, but the p-value denotes it is not significant enough to say it definitively.
3 | ConclusionBy looking at the democracy ratings, it seems that China hasn’t had the largest impact in changing regime characteristics.
This could be for a few reasons.
Most conspicuous is the fact that the dataset ends in 2014, just when China announced their absolutely massive and ambitious Belt and Road Initiative.
The BRI has quickly become China’s main program for investments abroad and is missing from our dataset.
The dataset also just misses President Xi Jinping’s assertive and arguably aggressive foreign policy, which he enacted shortly after assuming office in 2013.
Another is probably the lag effect of China’s aid and investments in these primarily developing countries.
Major transformations of governance have happened in recent years with the rise of populist leaders and nationalist sentiments.
Some of these have been in response to their government’s dependencies on China.
Could it be an unintended consequence or part of China’s grand strategy?Whatever the reasons, it will be very interesting to see what China does next.
Given the uncertainty caused by the trade war and a global economic slowdown, many stakeholders are probably enthusiastic and nervous about what the economic giant will do next.
Footnotes:¹Bluhm, Richard, Axel Dreher, Andreas Fuchs, Bradley Parks, Austin Strange, and Michael Tierney.
2018.
Connective Financing: Chinese Infrastructure Projects and the Diffusion of Economic Activity in Developing Countries.
AidData Working Paper #64.
Williamsburg, VA: AidData at William & Mary.
²Marshall, Monty G.
, Ted Robert Gurr, and Keith Jaggers.
2017.
“Polity IV Project: Political Regime Characteristics and Transitions, 1800–2016: Dataset Users’ Manual.
” Center for Systemic Peace: 1–82.
While you’re here…Please leave a comment, it could be about what you think about the article, other sources I should refer to or other datasets you think I should inquire.
Rio Rinaldi is a graduate political science student at Villanova University.
I am particularly interested in utilizing quantitative methods for my research.
Using R has helped me explore whatever dataset strikes me fancy and has helped me ask better questions.
Connect with me on LinkedIn or if you’re in the Philly area I would love to chat!.. More details