The phenomenon of Dutch disease and the impact of investments on the Portuguese economy

Dutch disease represents an unfavorable situation in the economy usually caused by the discovery of a new field and resulted in the production decline and as consequence stagnated economy. The phenomenon was firstly introduced as the explanation of the negative influence on the economy of the Netherlands due to the discovery of a significant number of gas deposits. Thus negative effect on the economy is usually determined by the rapid development of one particular sector and the subsequent decline in the others, which reflects mainly on the country’s export, import and GDP, and as well may lead to the rise of inflation and unemployment rates.

Despite the fact that natural resource boom is usually regarded as the major cause of the considering phenomenon, negative economic consequences may be explained by a massive inflow of foreign currency especially through foreign assistance and foreign direct investments. This assumption can be exemplified by Portuguese experience.

The researchers confirm that Portuguese economy was subjected to Dutch disease due to several reasons, among which the intensive development of natural resources is also mentioned. However the leading and principal reason of the phenomenon is believed to be foreign assistance as a result of the European Union integration and the accession of Portugal in it, thus foreign direct investment inflow is also considered to be a cause of Dutch disease.

The main aim of the paper was to investigate whether Portugal was still experiencing Dutch disease and to determine whether direct investments contributed to this process. Due to this purpose two separate models of systems of equations were constructed. The data were taken from the reliable sources, such as OECD database and FRED economic data. In accordance with the outlined objectives the first step represented an attempt to create a model including FDI as one of the regressors and estimate its impact on the country’s GDP and export. Assuming that GDP and export were dependent variables the following indicators were chosen for the models: LP- labor productivity, EMP – employment, EXPORT – export, national currency, IPI –industrial production index, GCONSE – government consumption expenditure, GF – gross fixed capital formation, FDI – foreign direct investment, CONSPR – private consumption expenditure, RTV –retail trade volume, IPI –industrial production index, WA – population of working age, NDI –net direct investment.

The obtained results showed the first model could not be estimated as a trustable one as the first equation included several unimportant variables (FDI was among them) and did not satisfy the third Gauss-Markov’s condition about the absence of autocorrelation. However, the second equation in the model could be regarded as adequate due to the R-squared and Durbin-Watson statistics figures, moreover with probability value of 0,0011 that was less than 0,05 value FDI should be treated as an important variable affecting the regressant in the equation. Thus it is possible to assume that FDI do have influence on the dependent variable and the decrease in investment leads to the increase in IPI and RTV for instance.

For the second model all dependent and independent variables remained the same and FDI variable was substituted with NDI one (net direct investment), moreover slight changes in time range were introduced due to the insufficiency of statistic data. The results are presented below:

Sample: 2000Q1 2010Q4

Included observations: 44

Total system (balanced) observations 88

	Coefficient	Std. Error	t-Statistic	Prob.
C(1)	-33154.63	1349.341	-24.57098	0.0000
C(2)	39348.69	1094.463	35.95250	0.0000
C(3)	0.007156	0.000217	33.03130	0.0000
C(4)	2.60E-08	1.75E-08	1.483882	0.1421
C(5)	9.784827	2.263351	4.323159	0.0000
C(6)	1.53E-07	5.11E-08	3.000343	0.0037
C(7)	9.43E-08	2.38E-08	3.954246	0.0002
C(8)	1.27E-08	4.23E-09	3.008836	0.0036
C(9)	7.07E+10	1.73E+10	4.077228	0.0001
C(10)	4.16E+08	94462918	4.401585	0.0000
C(11)	-0.083321	0.020056	-4.154488	0.0001
C(12)	84361809	24934544	3.383331	0.0012
C(13)	1.51E+10	4.71E+09	3.211202	0.0020
C(14)	77870747	11686765	6.663157	0.0000
C(15)	-13709.60	2233.949	-6.136934	0.0000
Determinant residual covariance		4.87E+19

Equation: GDP = C(1)+ C(2) *LP +C(3) *EMP + C(4)*EXPORT

+C(5)*IPI

+C(6)*GCONSE +C(7)*GF +C(8)* NDI

Instruments: LP EMP EXPORT IPI GCONSE GF NDI C

Observations: 44

R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat

0.999141  Mean dependent var  43736.12 0.998973  S.D. dependent var    1222.801 39.18036  Sum squared resid    55263.62 1.536006

Equation: EXPORT= C(9)+ C(10)*CONSPR +C(11)*NDI+

C(12)*RTV +

C(13)*LP+ C(14)*IPI +C(15)*WA

Instruments: CONSPR NDI RTV LP IPI WA CPI C

Observations: 44

R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat

0.983863  Mean dependent var  1.62E+10 0.981247  S.D. dependent var   1.57E+09 2.15E+08  Sum squared resid    1.71E+18 1.955921

In comparison with the previous model all the variables in both equations satisfy the condition of being less than 0,05 probability value except for the export variable in the first equation. However, it cannot be excluded from the model firstly due to the fact that it is one of the main components of GDP according to the formula and secondly, after running a t-Test it becomes clear that its t-value is less than t-critical (approximately 1,98) that supports the idea of the variable’s significance to the model.

Thus, from the interpretation of these model coefficients it can be seen that the increase in labor productivity, export, industrial production index and other independent variables from the first equation lead to the increase in NDI and consequently the overall increase in these variables leads to the changes in GDP. At the same time the increase in private consumption expenditure leads to the decrease in NDI and the following increase in retail trade volume, labor productivity, industrial production index and population of working age that leads to the change in export, hence the assumption that net direct investment can affect the dependent variable is confirmed.

R-squared figures for both equations are acceptable implying that equations are satisfying and reasonable. Turning to Durbin-Watson test results we cannot say surely whether autocorrelation is present or not, however in the second equation we can assume the absence of autocorrelation as figure is slightly less than 2. After running White test aimed at determining Heteroscedasticity we had to implement several adjustments to fight Heteroscedasticity in the first equation, thus we can speak about its absence in the model.

Summarizing all the above and taking into consideration the tests’ results we can come to the conclusion that investments do influence such key economic indicators as GDP and export. However, we cannot claim that the correlation among these factors accounts for the phenomenon of Dutch disease in Portuguese economy due to the nature of the model. Below the results of the tests do not allow us to treat the model as an infallible proof of the presence of Dutch disease in modern economy. Still we should consider the significance of net direct investments and at least to regard them as a premise of possible experience of this process.

List of sources

• 1.    João Sousa Andrade, Adelaide Duarte, The Fundamentals of the Portuguese Crisis // PANOECONOMICUS, 2, 2011. PP. 195-218.

• 2.    João Sousa Andrade, Adelaide Duarte, The Dutch Disease in the Portuguese Economy // Estudos do GEMF, #5, 2013.

• 3.    Zuzana Brinčíková, The Dutch Disease: An Overview // European Scientific Journal, August 2016.

• 4.    Трегуб И.В. Investment project risk analysis in the modern Russian economy // research in empirical international trade. – Slovenia: working papers. June. 2012.

• 5.    Трегуб И.В. Анализ модели инфляции с применением эконометрических и имитационных методов // сборник трудов научно-практической конференции «эконометрические методы в исследовании глобальных экономических процессов», 29 октября 2013 г. Москва, МГИМО (у) МИД России, фонд развития МГИМО. М.: АНКИЛ, 2013. С.223-232

• 6.    Трегуб И.В. Математические модели динамики экономических систем – монография, М.: 2009.

«Экономика и социум» №2(33) 2017