FDI and Productivity Growth in Africa: A Comprehensive Study

Mrs S Fauzel PhD Student University of Mauritius Associate Prof Seetanah University of Mauritius Associate Prof Sanassee University of Mauritius

Presentation Outline Introduction Brief literature review Aims and objectives of the present study Data and model specification Results Conclusion

Introduction FDI has acquired considerable importance as a tool for the economic development of host countries and for accelerating their growth. Inward FDI has been evidenced to boost aggregate investment and the level of economic activity. Besides, FDI has numerous benefits which include employment creation, improved productivity, enhanced exports, technological and knowledge transfers.

Introduction Attracting foreign direct flows also ranks high on African countries agenda in view of the accompanying wide ranging benefits. With the advent of the global financial crisis, such flows to Africa have been constantly on the decrease. FDI though remains a crucial ingredient which facilitates the flow of capital

Brief literature Review FDI and technology transfer Hummelsand Stern (1994), Lall (1996), Van den Berg (2001), Blomstrom et al., 2000, Jordaan (2012) FDI, economic growth and productivity Abdulhamid et al (2011), Bengosand Sanchez-Robles (2003), Bende-Nabendeet al. (2002) Limited evidence: Aitken et al. (1997) and Saqib and al (2013) Hence, the growth effect of FDI does not win unanimous support. Several problems were identified in previous studies. The problems were mainly in the face of a crowding out effect on domestic investment, external vulnerability and dependence, a possible deterioration of the balance of payments as profits are repatriated, destructive competition of foreign affiliates with domestic firms and market-stealing effect .

Brief literature Review Although the literature is fraught with studies analyzing the impact of FDI and technology transfer at the micro level, it could therefore be argued that only few have so far investigated the relationship between FDI inflows and productivity growth at the macro level. In this regards the paper attempts to fill this gap and to come up with new evidences.

Paper Structure The paper is in two fold: FDI and productivity spillovers in Africa FDI in the manufacturing sector and productivity spillovers in Mauritius. Contrasting with previous empirical studies, this part of the paper uses a dynamic vector error correction model (VECM) to carry out the analysis.

Aims and Objectives Whether foreign investment contributes in augmenting total factor productivity of the host countries. Is there a two way causality relationship between FDI and TFP growth? What are the important determinants of FDI that can be identified in the model used? Analysing the extent to which FDI generate a crowding out effect on domestic firms as postulated by various studies.

Data and model specification TFPG = f (FDI, TG, OPNS, HC, CPI) TFPG : new dataset for TFP developed by United Nations Industrial Development Organization (UNIDO)-UNIDO World Productivity dataset FDI: The ratio of FDI to GDP is used as a proxy of foreign presence and accounts for country size. TG is measured as the difference between the GDP of a particular country and the average GDP of all remaining countries in the sample. OPNS the ratio of exports plus imports to GDP is used as proxy. HC: secondary school enrollment ratio as a proxy for human capital. CPI is used as a measure of inflation Panel data: SUB SAHARAN AFRICAN countries over 1980-2010

Preliminary Tests Stationary test Using the Im, Pesaran, and Shin (2003) panel unit root test, we found that the series are non-stationary at their level and stationary at their first difference at 5 per cent level of significance. This means that the series follow an I (1) process. we subsequently test for the presence of long run relationships among the variables. The results show the presence of co-integrating vector and we thus conclude that a long run relationship exists between FDI, human capital, inflation, openness, technological gap and TFP growth.

Estimation Issues Given the possibility of endogeneity and causality issues we used vector auto regressions (VAR) on panel data to enable us to consider the complex relationship that might exist between FDI and TFP growth. We specify a first order VAR model as follows: Where zt is a SIX variable vector (tfp, fdi/gdp, tg, hc, opns, cpi) and the variables are as defined previously. We use i to index countries and t to index time. The lowercase variables are the natural log of the respective uppercase variables

Estimation Issues By using the PVAR any feedback and indirect effects which might be present will also be detected within the system.

Results from the PVAR model Response to Responseof Constant tfpgt-1 fdit-1 hct-1 openst-1 cpit-1 tg t-1 TfpG 0.51 0.98*** 0.12** -0.39 0.17* -0.02** 0.008* Fdi -1.82 0.25* 0.60*** 0.51* 0.29* -0.02* 1.35* Hc 0.21* 0.06** 0.01 0.91*** -0.06 0.004 0.12 Opens 0.42* 0.12* 0.01* 0.014 0.87*** -0.01 -0.11 Cpi 4.16 4.64*** 0.04 -3.82** 0.68 0.45*** -1.14 Tg 0.13 0.07 -0.01 0.11 0.19* 0.03 1.06***

Further analysis of findings Referring to the FDI equation, it is observed that a reverse causation exists as well as TFP growth appears to be also a determinant of FDI. (NOTE) Also, FDI as a dependent variable is highly influenced by all the other control variables. Consequently, it is observed, in terms of magnitude, that past values of FDI, human capital, openness and technological gap are all important determinants of FDI. Such results provide insights as to the policies that a country should have in order to attract FDI in the long run.

FDI in manufacturing and TFPm in MAURITIUS

FDI in Mauritius Mauritius has been successful in attracting significant FDI mainlydue to the following: low-cost labour Efficient infrastructure Preferential access to large markets Sound legal system Political stability Government policies investors(Business facilitation Act) and a strong business environment with avibrant entrepreneurial culture. favourable to foreign

Inward FDI flows 2001-2011: Mauritius Economy 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Mauritius -26 32 62 11 42 230 357 396 267 443 320 Memorandum: comparator economies Malawi 60 40 66 108 52 72 92 9 60 140 56 Madagascar 93 61 95 95 86 295 773 1,169 1,066 860 907 Kenya 5 28 82 46 21 51 729 96 141 133 335 Zimbabwe 4 26 4 9 103 40 69 52 105 105 387 Source: UNCTAD's FDI/TNC database, available at: http://stats.unctad.org/fdi/.

Mauritius: sectorial distribution of FDI inflows, 2006-2011 20111 Description 2006 2007 2008 2009 2010 0.81 0.57 15.54 - - 5.99 Agriculture, forestry and fishing 5.78 8.41 5.16 14.78 2.01 1.83 Manufacturing Electricity, gas, steam and air conditioning supply 0.55 - - - 0.06 0.03 0.37 1.40 2.36 6.43 41.05 70.91 Construction Wholesale and retail trade; repair of motor vehicles 6.32 1.19 3.59 8.86 3.97 0.71 and motorcycles 0.42 - 0.49 0.29 3.49 0.12 Transportation and storage 44.13 99.08 46.82 56.32 26.57 19.61 Accommodation and food service activities 1.36 0.57 0.27 - 7.48 2.57 Information and communication 114.73 126.03 158.54 41.75 147.57 55.74 Financial and insurance activities Real estate activities 54.32 118.70 157.17 131.07 108.72 155.09 of which - IRS/RES/IHS 39.21 86.71 91.60 63.14 64.58 113.50 - - - - 12.84 7.35 Professional, scientific and technical activities 1.74 0.93 2.57 3.81 0.57 0.14 Education 0.07 0.90 4.16 4.42 86.80 - Human health and social work activities - - - - 1.96 0.10 Arts, entertainment and recreation 230.61 357.78 396.68 267.72 443.11 320.18 TOTAL

TFP In Manufacturing Sector 180 160 140 120 100 Real Output Labour Productivity 80 Capital Productivity Total Factor productivity 60 40 20 0

Regression Analysis: FDI in manufacturing and TFPm TFPGm = 0 + 1FDImt+ 2DInvtmt+ 3HCt + 4CPIt + 5INSTt + t Added varibles: Dinvtm for local investment in the manufacturing sector INST for an institutional variable in the form of tariff. Time series analysis: 1980-2010

VECTOR ERROR CORRECTION MODEL, VECM We first investigated the unit roots properties of the time series Each variable is I(1) and the Johansen co-integration test identifies the presence of co integration, and hence a long run equilibrium relationship among the variables has been established. the next step is to specify and estimate a VECM including the error correction term to investigate the dynamic nature of the model.

Long run estimates of the VECM Dependent Var : ln TFPm t-ratios ln LPm t-ratios FDIm 0.105650*** 1.702310 0.090065*** 1.02331 Dinvtm 0.182167* 6.003050 0.129348* 1.77231 Phc 6.918044 13.5539 5.25681 6.57461 Inf -0.151613** -4.23056 -0.101267* -4.28213 -0.10225* Tariff -0.140991* -0.67882 3.599000 Constant -35.58135 -27.43292

Short run results Error correction Model tfpm fdim dinvt phc inf tariff Constant 0.02 -9.53 -0.73 -0.13 -0.23 -0.23 tfpm t-1 0.38* 0.013* 1.24* 0.17* -14.50 0.78 fdim t-1 0.005* 0.19* 0.16* 0.01 -0.07 0.01* dinvt t-1 0.13** 0.31* 0.28* 0.018** 2.24 -0.17* phc t-1 -0.60 -17.60 -1.70 0.55* -4.92 1.31 inf t-1 -0.01** -1.50* -0.15** -0.01 -0.34* 0.02* tariff t-1 -0.06* -0.02* -0.04* 0.04 1.80 1.50 t-1 -0.12*** -0.98* -0.05*** -0.06*** -0.09*** -0.50 R2 0.53 0.64 0.55 0.85 0.49 0.67 DW stats 1.98 1.88 1.99 1.75 1.40 2.02

Conclusion FDI AND PRODUCTIVITY SPILLOVERS IN AFRICA Rigorous panel VAR procedures were employed mainly to examine this complex linkage between FDI and TFPG over the years 1980-2010. Indeed, although FDI induced a positive impact on TFP growth, openness was also seen to be an important determinant of TFP growth. The PVAR approach has also enable us to conclude that human capital, openness, TFP growth as well as high technological gap all together are important determinants of FDI for the sample of countries used.

Conclusion FDI is an important ingredient for TFP growth in Africa. Results from the analysis indicated the presence of a bi- directional causality between TFP growth and foreign direct investment. This result is in contrast with some present studies.

Conclusion FDI IN MANUFACTURING SECTOR AND PRODUCTIVITY SPILLOVERS IN MAURITIUS FDI in manufacturing sector boost TFP of the sector both in the SR and LR but the effect is more important in the LR showing that foreign capital takes time to have its full effect on the economy. FDI in manufacturing also results in a crowding in effect on domestic firms which is in contrast with many studies. [limited evidence of the market stealing effect] Lastly, a Bi-causal relationship between productivity and FDI and productivity and domestic investment has been found showing another important feedback and dynamic effect.

THANK YOU