Foreign Direct Investment and Growth in Developing Economies:
The Mediating Role of the 2008 Financial Crisis
Andrei-Mihnea Savu BSc Thesis Economics & Finance Student number: 10774076 Academic Year: 2017-2018 Supervisor: Magdalena Rola-Janicka Semester 2, periods 2 & 3 Submitted: June 26th 2018 Faculty of Economics and Business
Statement of Originality
This document is written by Student Andrei-Mihnea Savu who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of Contents
Abstract ………...… 4
I. Introduction ………....…. 4
II. Literature Review ………...…….. 6
A. Relationship between FDI and economic growth ……….. 6
B. Relationship between FDI and the crisis …………...……….... 7
C. Hypothesis ……….. 8
III. Methodology ………..…….…… 8
A. Model Specifications ………. 9
IV. Results and Interpretation ………...………...………... 11
A. Simple Regression ………...……….……... 11
B. Main Regression ……….………...…….. 12
C. Relation to past research ……….. 14
V. Conclusion ……… 14
A. Summary ……….. 15
B. Limitations and suggestions for future research ……….. 15
Bibliography ………...……….. 17
Abstract
Foreign Direct Investment (FDI) is a powerful mechanism through which developing countries can get extra funding in order to help them progress through the advancement of emerging markets. However, while it being seemingly beneficial to host countries many researches have found that if used under the wrong circumstances it can be damaging to an economy. This paper aims to help further the understanding of FDI and its behaviour by seeing how the relationship between FDI and economic growth is affected by a financial crisis. By looking at data from 19 developing economies over the period 2004 to 2015 results show that the crisis had a positive impact on the relationship at hand.
I. Introduction
International transfers are important aspects of the global economy. One of the more important features of said transfers is Foreign Direct Investment (FDI). Playing an important role in developing as well as developed countries it has sparked a lot of interest for economists to research the subject.
According to OECD the level of FDI has risen over time and helped emerging markets develop and small businesses prosper, it has gone from around 600 billion dollars in 2004 to 1.23 trillion dollars in 2014. Moreover, developing countries are incentivised to attract FDI flows into their economy, as not only can it generate short-term profits due to the influx of capital, but also gains in the long-term gains as it helps emerging markets present in the economy grow (Iamsiraroj, 2016). Other benefits which FDI can bring to a developing economy consist in things such as jobs, specialized skills in the field of interest, better technology and overall better resources from developed countries (Gui-Diby, 2014).
On the other hand, while the level of FDI has increased over the years and developing economies have mostly stood to profit from the inflows they received, debates over whether FDI can be damaging to a economy are still present in recent studies on the subject. One of the main arguments against FDI is that it can sacrifice long-term profits from short-term ones, as profits from the emerging markets could be contained in their economy, instead a part of them will go back to the outside investors. Another argument that ties into that is that it can have a negative impact on local businesses, having an industry giant from another country monopolize markets in the host country (Zaman, & Valentina, 2012). Overall the impact which FDI has on a country is still highly questioned, and numerous studies have aimed to find a relationship between FDI and a countries growth.
In the scope of establishing a relationship between economic growth and FDI many researchers have done studies to understand the behaviour of FDI. From localized analysis such as the one conducted by Zaman and Vaslie (2012) who worked on the macroeconomic impact which FDI had in Romania, to
the work of Sasi Iamsiraroj (2016) which looks at the effect FDI has on 124 different countries. This paper works off these past studies to help further understand how FDI affects economic growth under certain circumstances. In order to do that our approach will be to analyze the interaction between FDI and economic growth in relation to a specific event in time.
In late 2008 the world was struck by what is now known as the worst financial crisis since the great depression. In the wake of this event countries around the world were subject to drastic decreases in output, employment and trade, with factors such as GDP in industrialized countries and real GDP growth in emerging economies dropping by 4.5 percent and 8.4 percent respectively (Laura, & Maggie, 2010, p. 2). In light of that event many economists have started analyzing the macro patterns and causes of the crisis. Studies such as the one conducted by Alfaro and Chen (2012) have gone to analyze how the financial crisis affected FDI, exploring different channels through which FDI affects performance and how the crisis impacted those channels. Vintila (2010) has gone to analyze how different crisis affected the level of FDI over time.
This paper builds on these past researches by looking at how a financial crisis impacts the relationship between FDI and economic growth. In order to do that historical data has been gathered on 19 different developing economies on the period 2004-2015. The scope of this paper is twofold, firstly we want to see how a period of economic distress, in this case the 2008 global financial crisis, affects the level of FDI in developing economies. Secondly we want to see if there is a change in the effect which FDI has on economic growth before and after the crisis . We can break that down into the following two questions: “How did the 2008 financial crisis affect Foreign Direct Investment flows in developing economies?”, and: “How did the relationship between Foreign Direct Investment and economic growth change following the crisis?”
In order to answer the above questions the main variables used will be defined. FDI represents the net inflows and outflows of a country which consists of either buying controlling shares in a business from another country or by opening a branch of a pre existing local business in another country. Economic growth is measured by GDP growth, and the financial crisis is a dummy variable which is equal to 0 preceding the year 2008 and equal to 1 after that. The other control variables will be discussed in greater detail in the third section of the paper. The regressions will be made using panel data on the 19 available countries, the sample was chosen purely on data availability and the condition that it must be a developing economy.
When going into the statistical analysis I looked at the interaction term between FDI and the financial crisis. Initially, based on what past studies, I expected a negative impact following the crisis, however the results show that after the financial crisis the effect which FDI had on economic growth has
The paper is outlined in the following way: The following section will review related studies and their main findings will be outlined and discussed. The third section will go over the model used and its specifications, as well as presenting the variables used and discussing their relevance. The fourth section will present the obtained results and their interpretation will be outlined. The final section of the paper will contain a summary and present the conclusion, as well as talking about limitations and suggestions for further research.
II. Literature Review
Research related to FDI and how it impacts various factors is plentiful, on the other hand studies analyzing the relationship between FDI and a financial crisis are scarce. However, studies that were made provide useful information about the topic at hand. The following section will address the literature written on the main aspects of the paper. Firstly, the relationship between foreign direct investment and economic growth will be discussed, after that the findings about how the 2008 global financial crisis affected FDI. Lastly the hypothesis of the paper will be presented and discussed.
II.a. Relationship between FDI and economic growth
When it comes to studies about the relationship between FDI and economic growth many seem to reach a common consensus. FDI overall seems to be effective when it comes to stimulate economic growth, working as a mechanism for developing countries to get funding from outside sources. Gui-Diby (2014) has found a mostly positive relation between FDI and economic growth. He conducted his research on 50 African countries over the period 1980 to 2009. The way he approached the subject is by looking at FDI as a percentage of GDP at their current prices, whilst controlling for government
consumption, population size, secondary gross school enrollment ratio, and a proxy variable representing the domestic investment. Following his study he found that while FDI had a significantly positive impact on economic growth, it was only so for the time period 1995 to 2009, whilst being negative from 1980 to 1994.
The second relevant study I chose to look at is the one conducted by Zaman and Vasile (2012), which only chose to focus on the effects FDI has in one country. They sought to find the macroeconomic impact which FDI has in Romania over the period 1991 to 2011. One of the things they chose to look at is where the capital coming from FDI is allocated. They have found that only about 4.5% of the FDI flows go into the high-tech and scientific industries, while non-tradable industries and overall lower tech markets represent 43% of the share. After analysing the data they arrived at the conclusion that when
looking at the growth which FDI brings into an economy it is important to assess the drawbacks it bring with it, either it being a tradeoff between short-term and long-term profits or bringing local businesses at a disadvantage.
Sasi Iamasiraroj (2016) looks at the bi-directional relationship between FDI and growth, his study being based on 124 countries spanning the period of 1971 to 2010. He points out that economic growth is not only based on FDI, but also on an interaction between FDI and a group of capital indicators, more specifically domestic capital, human capital, and foreign capital including other factors such as labor, institutions and government policies. Results show that there is a two way relationship between the two variables, in the sense that foreign direct investment stimulates economic growth, which in turn makes it more appealing for outside businesses to invest in their economy further boosting FDI.
A report on Foreign Direct Investment for Development by OECD (2002) also highlights some drawbacks that FDI has on an economy. Their data shows that while FDI can have positive short-term effects on an economy through inflows from the outside, it might damage long-term growth by making it less accessible for local companies to establish. Summarized, research on the relationship between the two shows that FDI can be a good nexus for growth, however it should be monitored as under the wrong conditions it can end up being damaging to an economy.
II.b. Relationship between FDI and the crisis
Alfaro and Chen (2010) look at the relationship between FDI and Establishment Performance, while their main focus is understanding what role FDI plays when it comes to micro economic responses to the crisis. They look at three distinct mechanisms, which include production linkages, financial linkages, and multinational networks in order to understand that relationship. Their results indicate that FDI has a significant role when it comes to a countries response to a crisis.
Dornean, Isan, and Oanea (2012) also look at the relationship between the global financial crisis and the level of FDI. Unlike the other studies they chose to use FDI as the dependant variable while looking at how it is impacted by economic growth and the financial crisis. This research is relevant to our paper as we want to see if there is a change in the level of FDI in developing countries following the crisis, and this work helps us better understand the impact. They have found that there is a negative relation between the financial crisis and FDI, and a positive one between economic growth and FDI. This looks logical as a financial crisis would make investors more careful with their money, while economic growth would attract future FDI flows as discussed above in the study done by Sasi (2016).
Summarized, research on the relationship between the two shows that FDI can be a good nexus for growth, however it should be monitored as under the wrong conditions it can end up being damaging to an economy.
II.c. Hypothesis
By assessing past literature on the subject the following hypothesis are to be tested in this paper:
H1: The level of FDI present in developing economies will decrease after the financial crisis.
A reason as to why we expect the level of FDI to go in that direction is because periods of economic distress raise uncertainty, which has negative effects on investment (Inklaar, & Yang, 2012). Not only that but by looking at previous crises the level of FDI seems to decrease (Vintila, 2011). This can be because investors are not as willing to take risks when faced with a crisis.
H2: Following the 2008 global financial crisis the effect which FDI has on economic growth will decrease.
We expect this result mainly because of the uncertainty that is created with the crisis. Because investors will be more cautious they might be inclined to invest in safer projects with lower returns, and consequently slower economic growth. An argument that can be brought against this can be because investors will be more prudent with their investments, they might choose fewer bad projects overall, thus having a positive effect overall.
III. Methodology
This section of the paper is constructed in the following way. Firstly, the model used will be described. After that the variables used will be presented and their relevance to the model will be outlined. Following that the regressions of the paper will be discussed.
III.a Model Specifications
As mentioned in the previous section this paper aims to answer two questions, firstly it seeks to find out if the 2008 global financial crisis had an impact on the level of FDI in developing countries, and secondly if the effect which FDI has on economic growth is changed following the crisis. In order to answer the above questions a study based on the financial crisis was conducted. For this study quarterly data on 19 different developing economies was gathered. The main model of the paper uses economic growth as the dependant variable, which was taken from the World Bank database. Data on the variable FDI was taken from Trending Economics, and lastly data for the controls unemployment and inflation was gathered from Datastream. These variables will be discussed in depth later in this section.
Data was gathered on a total of 48 periods spanning from the beginning of 2004 to the end of 2015, with some countries missing certain data points, however there was enough information on them to warrant keeping them in the model. After the data was collected it was organized as a panel in an excel spreadsheet. Once the data was arranged the spreadsheet was important into the statistical analysis tool Stata. From there I used the tools offered by stata to declare the data as a panel dataset. Following that multiple regressions were performed, from which the most important ones are panel data with country and year fixed effects. Other regressions made were used to further test the validity of the main findings.
In order to answer the questions presented by the paper multiple variables were used. The main dependant variable of the paper is economic growth, which is measured as the average GDP growth taken quarterly. The reason for using economic growth as the dependent variable is because this paper wants to see how the effect of FDI is mediated following a crisis. As economic growth encompasses most
economic factors related to the development of a country it made sense to use it to try to explain the effect.
Secondly, Foreign Direct Investment is used as an explanatory variable as it is the main factor put into question by this paper. It is measured in the net amount of FDI inflows and outflows in a country, and the unit used is in millions of euro. A negative coefficient would mean that FDI is an impediment to economic growth in developing countries, while a positive one would mean that it helps improve the economic conditions. A dummy variable for before and after the crisis is also used, as the paper studies the relationship between FDI and economic growth in regards to the financial crisis. This variable will be equal to 0 preceding the year 2008 and will be equal to 1 afterwards. A negative coefficient here would indicate that economic growth slowed down in the aftermath of the crisis while a positive one would mean it speeds it up. The variable of interest in this paper is the interaction term between the two, namely FDIxCrisis, which is used to determine the interaction effect between foreign direct investment and the
financial crisis. A negative coefficient would mean the strength of the effect which FDI has on economic growth is lower after the crisis while a positive one would mean an augmentative impact.
Beyond that control variables were used in order to reduce the noise of the model. Unemployment is taken on a scale from 0% to 100% and it is measured by what percentage of the available working population is not employed at the time. The other control variable used is inflation and it is measured by the percentage price change of a basket of goods and services consumed by households.
Based on existing research we would expect a negative coefficient when it comes to the financial crisis as well as the interaction term, and a positive coefficient when it comes to the variable foreign direct investment, as explained in subsection II.c..
The following regressions are ran using the variables discussed above. Firstly, in order to answer the first question, if there is a change in the level of FDI after the crisis, a simple regression is ran, which is used to test the first hypothesis of the paper.
FDI it = ɑi + 1 Post Crisis + ɛit;
This regression is done by using country fixed effects as indicated by ɑ. The reason for doing so is we want to remove unobserved time-invariant country specific confounders from affecting the relationship.
The second regression, which is also the main regression of the paper, used to test the second hypothesis, is the following multiple regression:
Economic Growth it = it + 1 FDI it-1 + 2 Post Crisis it + 3 FDIxCrisis it + 4 Unemployment it + 5 Inflation it + Δit;
To go into more depth about this regression, it wants to see how economic growth is affected by the explanatory variables. For the variables FDI and FDIxCrisis lagged values were used. This decision was made because the effect FDI has on a period is not immediately integrated into the other factors, as the capital received needs to be used to have an effect. On top of running country fixed effects like we did in the first regression, I also ran time fixed effects in order to eliminate the common trends found between all units of observation.
captures individual and timed fixed effects while ɛand Δ are error terms used to capture any random effects and residual unexplained variation.
This regression is ran two times, once on the period 2004 to 2015, and a second time excluding the years 2008 and 2009 when the crisis was at its peak. This is done in order to see if the change in the
effect of FDI is heavily influenced by the period of major distress following the crisis, or if the crisis changed the way FDI is used, independent of the crisis period.
IV. Results and Interpretation
This section will be organized in the following way: Firstly, in subsection IV.a the outputs of the simple regression will be presented and the interpretation will be presented. Section IV.b will have the results of the main regression tabulated and discussed. Subsequently, section IV.c will contain a detailed evaluation of the results gathered and will be put in contrast with the theoretical finding of the literature discussed in section II.
IV.a. Simple regression
Table 1 - FDI regressed on Crisis
Country Fixed Effects No restrictions
Post Crisis 262.5058***
(76.0066)
262.5078*** (75.9622)
R2 0.0046 0.0046
The above table shows the results of regressing the dependant variable FDI on the dummy variable crisis. It is a panel data regression on a sample of 19 countries on the period 2004 to 2015. The first regression is ran using the fixed effect specification while the second one has no restrictions put in place.
The statistical significance are as followed: * mean significance at a 1% level, ** mean significance at the 5% level, and *** mean significance at the 10% level. A lack of an asterisk implies no statistical significance.
Standard errors are included in parentheses.
The full outputs are available in the Appendix of the paper.
The above regression seeks to answer the first question of the paper, if the crisis had an impact on the level of FDI present in developing economies. We can see that the result is significant meaning that the financial crisis did indeed change the level of FDI present in those countries. Moreover, we can observe that the crisis variable has a positive coefficient, meaning that following the crisis the level of FDI increased, going against the hypothesis presented in section II. However since this is a simple regression there are numerous other factors that could influence the level of FDI that we did not account for, thus we cannot conclude that this relationship holds under different circumstances.
IV.b. Main regression
Table 2 - Economic growth in relation to other factors
Country and Year Fixed Effects Country Fixed Effects No restrictions FDI -2.89e-06 (1.94e-06) -2.09e-06 (1.92e-06) -2.57e-06 (1.84e-06) Post Crisis -0.0206*** (0.0071) -0.0232*** (0.0035) -0.0227*** (0.0035) FDIxCrisis 3.54e-06** (1.68e-06) 3.27e-06** (1.67e-06) 3.35e-06** (1.66e-06) Inflation 0.0735* (0.0410) 0.0931*** (0.0664) 0.0882*** (0.0329) Unemployment -0.0497 (0.0689) 0.0378*** (0.0071) -0.0506 (0.0589) R2 0.0587 0.039 0.0427
The results above observe a total of 3 panel data regressions on a sample of 19 countries on the period 2004 to 2015. The first regression is ran using both year and country fixed effects to account for unobserved time-invariant confounders and in order to eliminate the common trends found between all units of observation. The following two regressions lift one of the restrictions at the time in order to observe how the effects would change.
The statistical significance are as followed: * mean significance at a 1% level, ** mean significance at the 5% level, and *** mean significance at the 10% level. A lack of an asterisk implies no statistical significance.
Standard errors are included in parentheses.
Table 3 - Economic growth in relation to other factors (2008/2009 excluded)
Country and Year Fixed Effects
Country Fixed Effects No restrictions
FDI -2.34e-06 (1.83e-06) -1.94e-06 (1.79e-06) -2.41e-06 (1.72e-06) Post Crisis -0.0214*** (0.0065) -0.0209*** (0.0033) -0.0205*** (0.0033) FDIxCrisis 3.40e-06** (1.56e-06) 3.09e-06** (1.54e-06) 3.16e-06** (1.54e-06) Inflation 0.0824* (0.0445) 0.0903** (0.0425) 0.0802** (0.0400) Unemployment -0.2178*** (0.0675) -0.2425*** (0.0069) 0.0479*** (0.0081) R2 0.0361 0.0361 0.0275
The results above follow the same methodology is the previous table, the difference being that the years 2008 and 2009 are excluded from the regression.
The full outputs are available in the Appendix.
Looking at the above regressions we can observe that the dummy variable crisis as well as the interaction term FDIxCrisis are constantly significant. The negative coefficient for the crisis variable is to be expected as it causes economic distress, and as a result economic growth slows down. However the positive coefficient for the interaction term comes as a surprise, as it indicated that FDI has a stronger effect on economic growth in the aftermath of the financial crisis. The initial hypothesis of the paper was that following the crisis the effect which FDI has on economic growth will be lower, thus expecting a negative coefficient. Possible explanations for this might be that the crisis made investors more attentive when monitoring which investments to pursue, or because in the wake of the crisis host countries started
allocating their outside funds more effectively. In short the crisis might have served as a signal for both investors and host countries to pay better attention how they manage the fundings.
While finding why the crisis seemingly had a positive impact on the relationship between FDI and economic growth is a interesting subject that warrants further discussion, it goes beyond the scope of this paper to find why this interaction turned out this way.
IV.c. Relation to past research
When looking at the results I got in contrast with past research performed on the subject there are some important things to adress. In most based on the past studies I would not expect to find a positive relationship between the crisis and FDI. Vintila (2011) has found that FDI levels have decreased by significant levels in the wake of the crisis. Dornean, Isan, and Oanea (2012) have also found a significant relation between the financial crisis and FDI, however he found that the crisis had a powerful negative effect on his studies countries. He explains that by saying that because of the magnitude of the crisis affected economies with high FDI outflows, which in turn damaged the economies benefiting from the inflows.
In summary many of the relevant researches have found a significant relationship between FDI and economic growth. Even the ones looking at the financial crisis as a factor have found significant results when it comes to the impact FDI has on economic growth. On the flip side the papers which looked at the subject found results that would indicate the crisis had negative implications when it comes to mediating the effect which FDI has through the crisis (Vintila, 2011). As mentioned above, this goes against the results obtained following my analysis, meaning that both the hypothesis of the paper are rejected. As to the reason why the results seem to be this way could be an interesting topic for further research.
V. Conclusion
This section is outlined in the following way. Firstly, a summary of the paper will be presented, following that the main findings and important points will be discussed. Subsequently a section talking about the limitations, as well as suggestions for further research will be presented.
V.a. Summary
This papers scope is a study regarding the relationship between foreign direct investment and economic growth, with the aim being to see if the relationship of the two change following the 2008 global financial crisis. The research can be summarized into two main questions, firstly “How did the 2008 financial crisis affect Foreign Direct Investment flows in developing economies?”, and secondly “How did the relationship between Foreign Direct Investment and economic growth change following the crisis?”.
In order to approach the questions a literature review on relevant researches has been done. From the past studies we are expected to believe that the crisis will have a negative impact on the relationship at hand. To test that I have gathered data from numerous databases on 19 different developing economies. After the data has been gathered it was imported into Stata through an excel spreadsheet. Once the data was in excel I used the functions available in the tool to analyse it, a panel data with country and year fixed effects regression was ran. Results have found a significant relationship between the interaction term FDIxCrisis and economic growth, not only that but going against what was expected the coefficient is a positive one, meaning that the financial crisis actually had a positive impact on how FDI functions in developing economies. This could be because investors are more attentive when choosing what
investments they will take, or because the host countries benefiting from FDI are doing a better job allocating resources they are given in order to attract future investments. Explaining why the relationship changes goes beyond the scope of this paper, however it would make an interesting subject for future researches.
V.b. Limitations and suggestions for future research
While I tired to make this research as throughout as possible there are multiple limitations which influenced the results. Firstly, the data availability played a big part in the paper, initially having a sample of 50 countries but having to drop 31 of them out due to too much missing information. Secondly, while the research has provided some interesting results, they have to be taken with scrutiny as there are various other factors for which I did not account.
Another major limitation of the research were the time constraints, as a study on this subject could take a lot more to properly capture more information and more accurate results. Given that I feel like given the time frame available the methodology and results of the paper are satisfactory.
growth, it does a poor job of explaining why that interaction behaves the way it does. Future papers could aim to find out why there seems to be a positive interaction between the two. Besides that the same research could be done again while accounting for more factors that could be relevant to the subject.
Bibliography
Alfaro, L., Chen, M. (2010). Surviving the Global Financial Crisis: Foreign Direct Investment and Establishment Performance. Harvard Business School BGIE, Unit Working Paper
No. 10-110.
Dornean, A., Isan, V., Oanea, D.C. (2012). The Impact of the Recent Global Crisis on Foreign Direct Investment. Evidence from Central and Eastern European Countries. Procedia
Economics and Finance, 3, 1012-1017.
Gui-Diby, S.L. (2014). Impact of foreign direct investments on economic growth in Africa: Evidence from three decades of panel data analyses. Research in Economics, 68, 248-256.
Iamsiraroj, S. (2016). The foreign direct investment - economic growth nexus. International Review of
Economics & Finance, 42, 116-133.
Inklaar, R., Yang, J. (2012). The impact of financial crises and tolerance for uncertainty. Journal of
Development Economics, 97, 466-480.
Valentina, V., Zaman, G. (2012). Macroeconomic impact of FDI in Romania. Procedia
Economics and Finance, 3, 3-11.
Vintila, D.M. (2011). Foreign Direct Investments During Financial Crises. The Academy of
Economic Studies Bucharest, 41-45.
OECD Overview (2002). Foreign Direct Investment for Development, Maximising Benefits, Minimising costs. Retrieved from: https://www.oecd.org/investment/investmentfordevelop ment/1959815.pdf
Appendix
Figure 2 - Panel Data regression with country fixed effects
Figure 3 - Panel Data regression
Figure 5 - Panel Data regression with country fixed effects (2008/2009 excluded)
Figure 8 - FDI regressed on crisis with country fixed effects List of Countries - Albania - Armenia - Azerbaijan - Bolivia - Bulgaria - Colombia - Croatia - Ecuador - Egypt - Kazakhstan - Mexico
- Morocco - Peru - Philippines - Romania - Sri Lanka - Thailand - Venezuela
