Somalia. Afghanistan. Syria. These states are considerably the front runners for failed governance in the modern world, as they are countries which advance the threat of violence and insecurity to their neighbors, and whose very monikers evoke images of anarchy and chaos. The leading security interest of governments in the global community today is the control of autonomous territories (areas which have no influence of any legitimate central government). These territories are virtually independent, without authority or rule of law. The existence of such territories without the presence of state governance have enabled violent non-state actors such as terrorists to act openly without fear of restraint. There has been many efforts to control autonomous territories in the international community, for example governments have provided security assistance and peacekeeping operations. However with continued conflict and new coming cases of unstable countries like South Sudan and Libya, it seems the reactive responses might need to switch to more sustainable and proactive policies. With the growing number of countries flirting on the brink of state collapse, it begs the question of how can we prevent this phenomena at its core?
Analyzing the characteristics of weakening states, it seems that extreme poverty, resource scarcity, and general economic insecurity are a common precursor for state collapse, aside from the other more visible triggers of civil war and violent conflict. Focusing on economic weakness, I ask the question of whether a strengthened economy and robust industry can prevent state collapse in this blog. Specifically, I am interested in understanding whether a measurement of economic strength called foreign direct investment (FDI) can reduce a state’s level of fragility. It is my hypothesis that foreign direct investment inflows into a state, will reduce that state’s fragility.
Method of Analysis
I will conduct a statistical analysis where I will study the effects of
my variable of interest, “Fragility”, on the other variable of interest
which is FDI net inflows[1]. In order to prevent misrepresentation of
the effects, I will control for the effects of other factors that could
determine fragility including GDP[2], the regulatory quality of a
state, and the natural resource rents that are gained from production
and extraction.
Foreign direct investment can be described as private or multinational firms that create the operation of production within a country, such as the overseas factories you hear about in the news. Private firms setting up shop within a country can bring more than just jobs, they build schools, hospitals, learning centers and institutes. This is because in order to receive contracts and licenses from the foreign governments, firms give something more in return than just tariff collection. Such investment can have a greater impact on poverty than other charitable forms food aid and assistance. Scholars Faust, Grävingholt, & Ziaja studied the relationship of foreign aid and fragility, asserting that a kind of development assistance such as that provided by bilateral assistance and aid agencies is necessary to addressing the issue of fragility. However it seems the kind of assistance they study is limited in that many aid agencies act by cutting checks and just throwing money at the problem. FDI on the other hand can promote development by providing infrastructure as well as social assistance in the form of jobs, schools, etc. The impact FDI has can effectively bolster economies. Literature on violent non-state actors in the developing world present a common policy prescription of creating alternative opportunities for employment (CLEEN, 2014; LeVan, 2013; Omitola, 2012). Many young men in autonomous territories such as the Sahel, forests of Colombia, or Middle-East, have joined mercenaries to engage in covert and illegal causes. The state cannot promote alternatives in these areas and have weakened to the point that they are ineffective, which is why foreign private investment can provide the resources for sustained economic activity.

There has been a semantic shift in the literature where scholars and professionals have now discontinued the term “failed states” to define these countries, opting for the more inclusive term of “fragile”(Keister, 2014). The former term implied that the state has ended and no other authority structure might exist, and many people saw this definition of doing more harm than intended by discounting the agency and presence of more unconventional governance structures. I use data on fragility from the Fund for Peace Fragility Index. The index ranking of 1 means that the state is the most fragile in the world. The index ranking of 178 means that the state is the least fragile in the world, and on a sustainable track.
GDP per capita, purchasing power parity, indicates the economic strength of a country, as well as its human development. Higher GDP (per capita PPP) can mean there is high quality of living and low unemployment Regulatory quality measures the strength of regulatory framework and legal provisions within a state. Countries with strong regulatory quality may have better governance, and as such are less fragile and could also attract more investment. This variable is a percentile rank among all countries, and ranges from 0 ( the lowest or worst quality) to 100 (the highest or best quality)[3]. Natural resource rents are the sum of oil rents, natural gas rents, coal rents (hard and soft), mineral rents, and forest rents (World Bank Group, Indicators). “Rents” is an economic term use to represent capital returns. This variable is measured as a % of the countries' GDP. A large number indicates there is resource abundance and profitability. Controlling for these variables will allow us to isolate the effect of FDI on fragility.
Findings
Unfortunately, the graph clearly indicates that there is no relationship between our two variables of interest. In the regression model below, the high R-squared value of 0.78 shows that the model was robust and the control variables capture much of the explanatory power of Fragility.
Conclusion
The findings indicated that there was not a strong relationship between FDI and Fragility. I beleive this is because FDI data and method of analysis were based on a snapshot in time, if I were to include FDI data over time time, either by lagging the variable of FDI or including a time series analysis it may show a stronger relationship with a state’s fragility, because the effect would be fully captured. It seems the variables I chose where robust and in a future study I would only change the measurement I used of FDI[4].
Appendix
Works Cited
Of the 178 total countries in this analysis, there is an even distribution of highly fragile and very sustainable counties. The tables below gives examples of fragile states, as well as strong ones. South Sudan and Somalia are in the top two, however they are not a part of the lowest FDI recipients. FDI was evenly distributed as well and was not concentrated to particular countries or regions. However it Europe is a region dominated the list of strong and sustainable states as well as states with the lowest FDI percentages. High FDI inflows is not correlated to high fragility.
In analyzing the relationship between fragility and FDI, it seems there is no real correlation. Countries that are very stable and sustainable (like European countries) exhibit varying levels of FDI inflows. Countries that are extremely fragile also experience varying levels of FDI inflows. The graph below shows the of a relationship, illustrated by the ambiguous plot of data units. The Lowess line illustrates the general direction and form of the observations. The fitted values illustrate the predicted line of best fit based on the data. Both lines are flat and horizontal, which further indicates no correlation in the effect of Fragility on FDI.
Unfortunately, the graph clearly indicates that there is no relationship between our two variables of interest. In the regression model below, the high R-squared value of 0.78 shows that the model was robust and the control variables capture much of the explanatory power of Fragility.
The regression model shows the strength the variables have to explaining a state's fragility. Although there is no statistical significance at the 0.05 level amongst the variables, as shown by the asterisks, the R-squared value of 0.78 is very high which suggests that much of the variation in fragility can be explained by the variables in the model. Natural resource rents did the best job at explaining a state’s fragility, because its correlation was statistically significant at the 0.01 level. Regulatory quality and GDP also did a better job at explaining a state’s fragility than FDI.
The findings indicated that there was not a strong relationship between FDI and Fragility. I beleive this is because FDI data and method of analysis were based on a snapshot in time, if I were to include FDI data over time time, either by lagging the variable of FDI or including a time series analysis it may show a stronger relationship with a state’s fragility, because the effect would be fully captured. It seems the variables I chose where robust and in a future study I would only change the measurement I used of FDI[4].
What remains seen is the fact that a high concentration of fragile states are developing nations. FDI can serve as a great boon to such developing countries in the global south, which hold the most fragile states. Historically, Europe experienced its own period of hyperactive capital flows and investment (intra-regionally) which advanced their economic modernization and security. This type of investment is even more greatly needed in developing countries where capital and resources are scarce. Capital investments flowing from the developed countries into the developing must be encouraged in order to ensure a brighter and promising future for weak states.
Appendix
[1] Foreign direct investment, net inflows (% of GDP averaged for the 2010-2014 period). Foreign direct investment can indicate that the level of ‘risk’ perceived by foreign investors is low, indicating a strong state. This would imply that states who have high fragility would have low FDI, and states with low fragility would have high FDI. But examining cases with high FDI’s despite incredible risk would eliminate that question of causality and focus on which came first, investment or lack of fragility.
[2] GDP per capita (2014 US$) GDP Per Capita Purchasing Power Parity incorporates the relative value of different currencies, as well as gives a view of the productivity of a country relative to its population. High GDP per capita can indicate higher standard of living. Controlling for this variable would eliminate the confounding factor of economic growth and human development, since both influence a state’s fragility and attraction of investment. High amount can also indicate the size of the consumer market/demand for potential sellers market their goods.
[3] Regulatory Framework is a big pull for investment, this is often established in developing countries by International Organizations and NGOS, or developmental assistance from aid investment such as MCC, OPIC, USAID. These organizations seek to empower the technocrats within government administrations. Controlling for this variable would eliminate the confounding factor of Aid assistance, and good governance being a pull for foreign direct investment. “the Regulatory Quality index reflects perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development” (The Worldwide Governance Indicators (WGI) 2013).
[4 ] Note in the appendix collinearity tables below that the variables are not highly correlated (with perhaps the exception of regulatory quality and GDP (Per Cap, PPP) where the r of the product is 0.69). Note further that the variation inflation factor (VIF) indicates that the tolerance of all the variables are very high, and that it falls below the “rule of thumb” of 10. These tests indicate that there is no evidence of multicollinearity. (For more information see “Multicollinearity” by Richard Williams, Notre Dame, https://www3.nd.edu/~rwilliam/stats2/l11.pdf)
Works Cited
CLEEN Foundation. 2014. Youths, Radicalisation and Affiliation with Insurgent Groups in
Northern Nigeria. Lagos and Abuja: CLEEN Foundation.
Faust, Jörg, Grävingholt, Jörn and Ziaja, Sebastian. “Foreign aid and the fragile consensus on state
fragility” German Development Institute. 2013.
Keister, Jennifer “The Illusion of Chaos: Why Ungoverned Spaces Aren’t Ungoverned, and Why
That Matters” Cato Institute. 2014.
That Matters” Cato Institute. 2014.
LeVan, A. Carl. 2013. “Sectarian Rebellions in Post-Transition Nigeria Compared.” Journal
of Intervention and Statebuilding (2013): 1-18.
Omitola, Bolaji. 2012. “Terrorism and the Nigerian Federation.” African Security Review 21
(4):4-16.
Sources of Data
Fund for Peace, Fragility Index, 2014
Fund for Peace, Fragility Index, 2014
<http://ffp.statesindex.org/rankings-2014>
Fragility
World Bank Group,
World Bank Indicators, 2014
<http://data.worldbank.org/indicator/BX.KLT.DINV.WD.GD.ZS>
FDI & GDPPPP & Natural Resource Rents
FDI & GDPPPP & Natural Resource Rents
World Bank Group,
Worldwide Governance Indicators (Regulatory Framework, 2013
<http://info.worldbank.org/governance/wgi/index.aspx#reports>
Regulatory Quality
The research questions is clear and we can identify easily the key DV and IV. However, I am cautious about the explanatory variables. Even though it was explained in detail the limitations and the bias, I think it is important to identify whether the model shows problems of multicollinearity. It seems that it may be a problem of measurement since the fragility index already may have a component of conflict commodities. It would be important to add a correlation test among the control variables to solve this inconvenience. Another assumption mentioned on the post was the time order issue. regarding this limitation I personal believe that explaining more in detail the history track, you can support that FDI comes first than State fragility if you support your research on qualitative studies.
ReplyDeleteAdditionally, it would be useful adding a paragraph with findings and conclusions, thus you make sure to respond your question.
The final comment is related with the formatting of the table and the graph.
You do a good job explaining your research question, variables and how everything is measured etc.
ReplyDeleteHowever, instead of so much technical jargon and explanation in the main body of your post, why not move that to the footnotes? I think some more context would be really nice. What other studies are out there like this? Where does yours fit in the field? Is there a specific example/ case study you can cite that lines up with your research?
It's also helpful to succinctly explain whether or not you rejected the null hypothesis, and break down why in layman's terms. What exactly did your findings mean in regards to your hypothesis?
There are some typos you probably want to fix. And the graphs and charts could be a lot neater and pretty looking.
The main thing your post would benefit from is some context explaining why this is important, and citing an example that supports your hypothesis. It's an interesting topic and I know I'm at least interested in knowing more about it and what others in the field have to say.
I really appreciated how clearly the research question was stated and how the IVs and DV were immediately identified, which made it much easier to understand the rest of the post. This is obviously an extremely important question, and it is important to find an answer to it.
ReplyDeleteAs for the variables, it would have been nice to have some explanation of the grading of the fragility index. What factors are considered in making that index? Also, it would be great if you could mention how many points the index is made up of. Later in the regression table you mention that 1 percent FDI increase leads to 90 point increase on the fragility scale, but we don’t know how many points this is out of. As for FDI, is that total flows of FDI? In that case, there’s no control for country size, which would make a big difference in terms of FDI flows. As for the conflict commodity variable, I was a bit confused about whether the presence of five conflict commodities really translates to a higher volume. Isn’t it possible that one country could produce relatively small amounts of five conflict commodities while another country could produce large amounts of just one?
It might be hard to draw a conclusion from this, seeing as it appears that there is not a statistically significant connection between FDI and state fragility, but your final model does indicate a very high R-squared value, which means it has a good deal of explanatory power. In that case it would be nice to have an explanation of how the other variables do a better job of explaining state fragility than FDI does.