BACHELOR’S THESIS
AN EMPIRICAL ANALYSIS ON THE PORTUGUESE ECONOMIC
GROWTH AND FISCAL CONSOLIDATION
Written by
Iuri Willem van Eck dos Santos
10803254
Supervised by
Péter Foldvari
2
Statement of Originality
This document was written by Student [Iuri Willem van Eck dos Santos] who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is 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.
3 Table of Contents Page 1. Introduction 4 2. Literature Review 5 2.1. Concepts 6 2.1.1. Fiscal Consolidation 6 2.1.2. Economic Growth 6
2.2. Theoretical Approaches to Fiscal Consolidation 9 2.3. The Portuguese Case 11
3. Methodology 12
4. Data 13
5. Results and Findings 14 5.1 Results and Findings from [1] 14 5.2 Results and Findings from [2] 16
6. Limitations and Conclusions 17
4 1. Introduction
When the sovereign debt crisis hit Europe after the financial global crisis in 2008, countries such as Portugal, Ireland, Spain and Greece whose governments could not finance their daily activities had to be financially bailed out by other EU countries, the International Monetary Fund (IMF) and European Central Bank (ECB). This led to both economic and social consequences for the affected countries since the entities that lent the money demanded structural changes. The entities demanded these structural reforms in order for these rescued countries to economically recover and to maintain sustainable levels of debt. One of the main issues with the countries rescued was high and unsustainable public debt. Thus, with the aim of reducing their public debt, the structural reforms introduced were of fiscal consolidation nature. Fiscal consolidation has been a controversial topic as uncertainty remains on whether its success depends on the state of an economy, country-specific characteristics or the short and long-run consequences of austerity. Therefore, the aim of this paper is to analyze if fiscal consolidation measures can influence the convergence path to the steady-state level of income as described by the Solow model and to infer if fiscal consolidation has had an positive impact on economic growth in Portugal for the period 1975 to 2014. This analysis will be conducted by means of an empirical analysis basing the economic variables and growth as predicted by the Solow growth model and Cobb-Douglas production functions.
In the case of Portugal, the most recent fiscal consolidation process took place in March of 2010. This occured when Portuguese policy-makers went from passing a package of fiscal stimulus to fiscal strictness as a result of the world crisis and the beginning of the European Sovereign debt crisis (Caldas, 2012). Due to Portugal’s unsustainable existing public debt and the inability of its government to financing its daily obligations (i.e. senior pensions, salaries of public servants etc.), the Portuguese Government signed the Memoranda of Understanding in 2011 after months of debates with the troika (a board formed by economists representing the interest of other European Union countries, ECB and IMF). Portugal would receive a loan of €78bln (about 45% of GDP) that was conditional on a set of measures that had to be implemented to reduce the deficit and public debt (Cabral et al., 2013).
Another term used for fiscal consolidation is austerity. Austerity measures have been used as the go-to option by EU policy-makers when problems surge due to high public debt. One of the disadvantages of this type of policies is that they have been implemented independent of the state of the economy and even in the extreme cases of a recession. Economists remain divided in this subject, as some claim that the benefits and confidence that fiscal consolidation brings to the economy outweighs its necessary efforts whereas others hold a contrary belief. One of the main opponents of this method being implemented in a
5
recession period was John Keynes who in 1937 said the famous quote “The boom, not the slump, is the right time for austerity at the Treasury” (Jayadev & Konczal, 2010, p.1). The Keynesian view believes that fiscal consolidation is contractionary because, if the government reduces its expenses or increases taxes, then the aggregate demand will be reduced as well as income, particularly in the short term (McDermott & Wescott, 1996). The aggregate demand in the Keynesian view is calculated by simple IS-LM models, which support this rationale. This will result in a contraction of the economy to an even more undesirable state (Pescatori, Leigh & Guajardo, 2011). Other scholars defend that there is an expansionary effect of austerity, which can translate into positive growth of the economy. This has been based on the expectations theory and was first brought to light by Giavazzo and Pagano (1990). They conducted a research on the economic growth of two Europeans countries and noticed that fiscal consolidations affected individuals’ future expectations of taxes and government purchases and therefore, their consumptions are significantly affected by the expansionary expectations. By performing this research, one should be able to infer if the expansionary view predominantly explains how the Portuguese economy performed the last four decades.
The rest of this work continues as follows. First, an extended review on relevant literature on matters of fiscal consolidation, economic growth and the combination of both in Portugal is discussed in Section 2. Additionally, a description on the models used as bases for this research is explained in Section 2. Section 3 presents an analysis on the method used to obtain conclusions from this research. Section 4 describes the data used and explains how the collected data will be useful in conducting this investigation. Section 5 presents the results of the data analyses. In the sixth section, the results and findings obtained will be examined and discussed. Section 7 provides a retrospective evaluation on the shortcomings of both the data and the method used, and suggestions for future research. Finally, Section 8 offers the conclusions of this research and an assessment of the quality of the answer to the research question: “Did fiscal consolidation have a positive significant effect on the Portuguese economic growth?”
2. Literature Review
In order to investigate the aims of this paper, the concepts of fiscal consolidation and economic growth as well as the reason as to why it has been scrutinized will be explored in this section. A clarification on these concepts will be explained and previous literature of these concepts as well as a background information on the fiscal consolidation in Portugal will be reviewed in detail. Additionally, the foundation of the models and the models used to gather information on the effect of fiscal consolidation in Portugal will be described in this section.
6 2.1. Concepts
2.1.1. Fiscal consolidation
Fiscal consolidation, also known as fiscal adjustments, fiscal austerity or just simply austerity, is a set of measures implemented by a Government in order to reduce national debt, which is usually caused by excessive consecutives fiscal deficits. Fiscal deficit is a result of government expenses exceeding government revenues, which for the case of governments is mainly based on tax revenues due to its nature. Fiscal consolidation consists of either an increase in taxes, which will lead to an increase on the revenue side of the balance, or a decrease of government spending, which will lead to a decrease on the expense side of the public balance. Economists have identified two different approaches to fiscal consolidation, namely, the quantitative and the qualitative approach.
The qualitative approach recognizes the aims of policymakers’ austeritive consolidations by studying their speeches and activities. This is, by means of analysing their narrative, thus it is also known as the narrative approach. It is considered to be a qualitative approach if, after assessing policymakers’ statements (either verbal or written documents), deficit reduction nature goals are identified. The statements that are used to identify qualitative fiscal consolidations are based on historical budget discourses or reports that are mainly motivated by large public deficits rather than cyclical economic disturbances performed by either government officials or other individuals of influence on the matter (Kleis & Moessinger, 2016).
In contrast to this approach, Kleis and Moessinger (2016) describe the quantitative approach as using proxies of fiscal indicators as way of identifying fiscal consolidations instead of statements made by policymakers. As different scholars use different set of indicators, there are vast forms of quantitative fiscal consolidation. These indicators have, however, as a common ground been used as instruments to measure improvements and/or failures in variables interpreting debt of a country or primary budget balances. Dependent of the indicator chosen, a certain level of deviation is set, and this is used as a standard to qualify a fiscal adjustment sustained by a country or multiple countries under study. This will be the approach used in this paper to measure the impact of fiscal consolidation on economic growth in Portugal.
2.1.2. Economic growth
Economic growth can be generalized as the change in an economy between two periods. This difference can be either negative or positive, depending on different factors that have a significant impact on the growth of the economy. In this paper, it is going to be analyzed if fiscal consolidation measures, alongside other control variables, had a significant impact on the economic growth of Portugal. Several models have been constructed to study the
7
aggregate production of a country and its factors of production. The goal of this paper is to analyze the following expression and to measure the impact of government deficit on the dependent variable:
Economic Growth = f (Capital, Labour, Human Capital, Trade, Time-trend, Government Deficit)
One of the models that best describes the level of production, a measure which estimates economic growth of a country and examines how it changes over time, is the Solow growth model. The Solow growth model and its underlying variables have been studied various times by scholars. One of the most influential studies on this matter was performed by Mankiw, Romer and Weil (1992) who carried out an empirical analysis on the economic growth explained by the Solow model. The used terms and the relations between the factors of their analysis will be used as the core of this research. By deriving the underlying variables of this theory and adding a variable representing government deficit, one should be able to assess whether austerity has a positive impact on the growth of the economy. The Solow growth model defines capital (K) and labor (L) as inputs and factors of production (A) as exogenous. These three variables are used to predict the level of output (Y) at a given time (t) by the means of a Cobb-Douglas production function exemplified below:
𝑌(𝑡) = 𝐾(𝑡)𝑎(𝐴(𝑡)𝐿(𝑡))1−𝑎 (1)
From this equation, Mankiw et al. (1992) derive one of the versions of the augmented Solow growth model, which takes into the consideration of the accumulation of human capital. The variables are divided by A(t)L(t) making the predictions interpreted as per effective labour units. Then, its natural logarithms are used for growth analysis. Presented below are the basic expression in which the equation is constructed upon and the steps implemented by these scholars to obtain subsequent changes. Finally, an expression describing the relation between the variables of the Solow growth model and steady-state income level, in equilibrium is formulated. This expression will be used further in the present analysis to investigate the impact of fiscal consolidation on this convergence path to the steady state level of income. Below are the steps and equations used by Mankiw et al. (1992) to derive the expression that defines the steadyy-state level of income.
𝑌(𝑡) = 𝐾(𝑡)𝑎𝐻(𝑡)𝛽(𝐴(𝑡)𝐿(𝑡))1−𝑎−𝛽 , 0 < α + β < 1 (2)
8 ℎ(𝑡) = 𝑠ℎ𝑦(𝑡) − (𝑛 + 𝑔 + 𝛿)ℎ(𝑡) (3b) 𝑘∗= (𝑠𝑘1−𝛽𝑠ℎ𝛽 𝑛+𝑔+𝛿) 1 1−𝛼−𝛽 (4a) ℎ∗= (𝑠𝑘𝛼𝑠ℎ1−𝛼 𝑛+𝑔+𝛿) 1 1−𝛼−𝛽 (4b) ln [𝑌(𝑡) 𝐿(𝑡)] = 𝑙𝑛𝐴(0) + 𝑔𝑡 − 𝛼+𝛽 1−𝛼−𝛽ln(𝑛 + 𝑔 + 𝛿) + 𝛼 1−𝛼−𝛽ln(𝑠𝑘) + 𝛽 1−𝛼−𝛽ln(𝑠ℎ) (5) Equations (3a) and (3b) describe the growth of the variables k and h per effective labor units. The steady-state expressions for these two variables are defined by equations (4a) and (4b), respectively. The expression which joins the first two formulas is specified by equation (5). From this, it can be deduced that the steady-state income per capita depends on the growth of the population and the accumulation of both human and physical capital labour. This is used as an indication for the convergence path of the Portuguese economy. Based on the analysis of Mankiw et al. (1992), the formula for convergence is presented below.
𝑑 ln(𝑦(𝑡))
𝑑𝑡 = 𝜆[ln(𝑦
∗) − ln(𝑦(𝑡))] (6)
where 𝜆 = (𝑛 + 𝑔 + 𝛿)(1 − 𝛼 − 𝛽)
Substituting equations (2) to (5) into equation (6), a simple transformation constructs the following expression: ln(𝑦(𝑡)) − ln(𝑦(0)) = (1 − 𝑒−𝜆𝑡) 𝛼 1 − 𝛼 − 𝛽ln(𝑠𝑘) + (1 − 𝑒 −𝜆𝑡) 𝛽 1 − 𝛼 − 𝛽ln(𝑠ℎ) − (1 − 𝑒−𝜆𝑡) 𝛼 + 𝛽 1 − 𝛼 − 𝛽ln(𝑛 + 𝑔 + 𝛿) − (1 − 𝑒 −𝜆𝑡) ln(𝑦(0)) (7)
A variation deduced from equation (7) is used as a proxy to measure the effect of fiscal adjustments on the economic convergence of Portugal, which is presented by the first regression in Section 5. Section 3 provides an explanation on the how these variables are transformed as well as an interpretation of the transformed variables.
Around the seventies, a famous study about the restrictions of economic growth named “The Limits to Growth” was presented and brought to the public scope under the
9
famous Group of Rome (Nordhaus, 1992). The idea presented in this work suggests that economic growth was not endless, and at some point, economic growth would even stop at its steady-state due to the limitations of natural resources. This view is seen as the Limit-to-Growth and claims that economic growth stagnates at some point in time. Nordhaus (1992), who was one of the main opponents of this school of thought, found evidence that this was not the case. He further claimed that the scarcity of these reserves did not stop economic growth but it did have a slowdown effect on the naturally increasing growth. This can be considered as a “dragged” effect on growth. By following his line of thought, the second regression of this analysis is constructed in order to examine if fiscal consolidation acted as a dragged effect on the Portuguese economic growth from 1975 to 2014. Based on on a derivation of the Cobb-Douglas production function for Portugal, this regression model is presented in Section 5.
2.2. Theoretical approaches to Fiscal consolidation
Various economists have analyzed the short and long-term consequences of fiscal consolidation on economic growth. It can be concluded that this can be separated into two different theoretical visions; the expansionary fiscal consolidation adjustments, also referred to as the expectations view, and the contractionary fiscal consolidation, also known as the neoclassical view.
Felstein (1982) first investigated the Ricardian equivalence, a theory that bases the expansionary approach of fiscal consolidation. He studied whether government deficits had an impact on aggregate demand or if it was negligible. The Ricardian rationale, or as Felstein refers to it; the “pre-Ricardian” equivalence, is defined in his work as being the adjustment that consumers do when there is a change on taxes or governments spending. He inferred that households adjust their consumption as they perceive i.e. a tax decrease, not as an actual increase in their disposable income but has a “postponement” of this tax burden because, eventually, the government will have to finance the prior tax reduction with a tax increase. In his research, Felstein also mentioned the fact that the general public as difficulties in recognizing whether the tax change is of brief or lasting nature and therefore, households expectations are subjected to bias, which in itself can make an impact on the outcome (positive or negative) of the implementation of the fiscal adjustment. This is due to the effect that expectations have on the economy and its relevant factors (i.e. interest rates, propensity to consume, etc.). His work essentially evaluates to what extend individuals expectations correspond to and therefore, “create” its own future, due to the nature of the Ricardian equivalence.
Giavazzo and Pagano (1990) followed this idea on expansionary effects of fiscal consolidation and first did an empirical research and analyzed intensively two different
10
countries, in what they considered to be the most “severe” cases in Europe at the time, Denmark and Ireland. Their conclusions were key to this theoretical view of expansionary fiscal consolidation as they demonstrate that the expectations expansionary effects of budget (deficit) cuts have significant prevailed in, at least, these two countries. Their interest was to evaluate which of the perspectives, contractionary or expansionary, would prevail in these two samples. They found that the expectation theory, which can be related to the Ricardian equivalence, actually surpassed the Keynesian view in Denmark. Additionally, they stated that the expectations of future changes, in taxes as well as government expenditure, led to a Danish consumption increase. This is contrary to the Keynesian view predicted contractionary consequence of fiscal consolidation performed during an “economic slump”, which was considered the situation of these two countries at the time. In the Irish study, the Keynesian effect did correctly predict the contractionary effects resulting from a tax increase and budget cuts, although it was not as severe as anticipated. The authors discuss that this happened due to liquidity restrictions existing in Ireland at the time, and therefore the Irish real consumption reduced by 7% when the first fiscal consolidations were implemented in 1982. Giavazzo and Pagano also differentiated the channels of the economy that converted into these two different outcomes, namely: tax channel, inflation channel, interest rate channel and substitution channel. Forecasting effects based on Giavazzo and Pagano (1990) paper, of the utility of this channels into successful fiscal consolidation implementation can be a difficult task because Portugal, as well as the two above mentioned countries, nowadays is part of the European Monetary Union (EMU). Due to the single currency, these countries have theoretically no power on its interest rate, inflation rate or exchange rate as it is set by the European Central Bank. Therefore, in order to better understand the effects of this economic channels and their influence on the consequences of fiscal consolidation executed in countries which this applies, this has tobe done according to the European values for these variables and from 2002, period from which adoption of the euro currency started, onwards.
Following this disruptive information, the investigation on the matter naturally increased in the following years. Alesina and Perotti (1995) concluded, after their cross-country analysis, that fiscal adjustments are more successful when there are expense reductions based instead of tax increase based. This is due to increase in competitiveness resulting from a government cut. This happens because, according to them, it sends a stronger message to the general public since it is harder to revert and therefore creates a “crowding-in effect”. This “crowding-in” effect is described in their paper by being the result of increased belief by investors on the success of the fiscal adjustment and so the interest rates will drop. This drop is explained by the risk premia fall, caused by the public perception that consolidation will take place and be successful so they will require less return since there is less risk involve, which will attract private investment. This increase on private investment
11
will then be larger than the decrease in aggregate demand caused by the fall in government expenses leading to a positive overall effect.
More recently during and in the aftermath of the world crisis of 2008, in a study conducted by Jadhav, Neelankavil and Andrews (2013) found that the austerity programs implemented in Eurozone countries did not achieve its goal of stimulating the economy supporting the Keynesian view. They state that the deficit reduction targets were set too high and that these were put in action before a sustained economic recovery was in place, which is, according to them, necessary for a successful austerity plan implementation. Because of this, the debt-to-GDP ratios, which were, according to the authors, the main goal of these policies, did not change as desired and it worsened these countries, already fragile, economic state. A divergent view was given by Alesina and Ardagna (2010) that defend that fiscal austerity was not prejudicial to the crisis that started in 2008 worldwide, but it actually helped countries recover from it. They assessed how fiscal policy affected the economy and investigated the fiscal stimuli and adjustments that occurred in the OECD countries, in specific, the difference on raising taxes against the lowering of government expenses. They found that the latter case usually leads to an aggravation of a recession much more than an adjustment on the tax increase does. Wachtmeister and Ödeen (2015) tried to infer if fiscal consolidation was helpful in the recovery process of Eurozone countries economy as a whole. They identified two different types of groups, the Baltic countries and the PIIGS (Portugal, Italy, Ireland, Greece, Spain,), where Portugal is included, and stated that austerity does not impact PIIGS countries as it does the Baltic countries due to the difference of cultures of both countries and different background settings, i.e. individuals in the Baltic area have been more used to live in harsh economic conditions. Therefore a deeper knowledge on the characteristics of a country is needed in order to forecast effects of fiscal consolidation correctly.
2.3. The Portuguese case
After the last big-scale fiscal adjustment episode in Portugal, a reflection started on whether the proposed “solutions” to the economic turbulence lived prior 2010 did not in fact only worsen the Portuguese scenario. In August of 2012, years after the world crisis of 2008, unemployment reached its record all-time high of 15,9 percent in Portugal, numbers that are specially alarming since young unemployment was higher than 30 per cent in that year (Gorjão, 2012).
Economists believe that instabilities and “unhealthy” economic practices started in Portugal years before. In order to be eligible for the entrance in the EMU, Portugal, had to meet, like every other EMU country, the economic requirements imposed at the Maastricht treaty that was signed in the beginning of 1992. Due to high public deficits, higher than seven percent of GDP in the beginning of the Nineties, a period of fiscal tighten followed so
12
that the entry was feasible and the convergence path begun between 1995 and 1996. This period of consolidation was characterized as of fiscal constraint as explained by (Cunha & Braz, 2007). After this period, Portugal went back to its “traditional” fiscal relaxing. This was only possible because of the contrary effects caused by the drop of interest payments, the slowing down consequences of disinflation and the belief given by the previous two years of tightening that the fiscal requirements were already met which boosted the economy (Cunha & Braz, 2007). While this was happening, the Government balance deteriorated which caused the Portuguese economy to go again to an unsustainable path of disproportionate accrued deficits and therefore, high public debt. By this work done by Cunha & Braz it can be inferred that Portugal has had a systematic problem with high public deficits that inevitably end up in excessive debt, thus fiscal consolidation has not been used efficiently to solve these problems and they persist through different periods. It is therefore, important to understand how these fiscal adjustments affect the growth in order to solve the puzzle of when to use it more efficiently.
3. Methodology
Based on the Solow model used by Mankiw et. al. (1992), two time-series regression models are constructed to assess the impact of fiscal consolidation on the economic growth in Portugal over the sample period of 1975 to 2014. In this paper, the variable government deficit is used as a proxy to represent fiscal consolidation. Firstly, the regression model [1] tests whether there is an impact of fiscal consolidation on the steady state level of income. In the second specification, regression model [2] tests the direct effect of fiscal consolidation on economic growth and if it acts as a “drag”. Both regressions are of time-series natureThe two regressions that investigate the relationship between fiscal consolidation and economic growth in Portugal are modeled as follows:
∆ ln(𝐺𝐷𝑃𝑡) = 𝛼0+ 𝛼1𝐿𝑛(𝐺𝐷𝑃𝑡−1) + 𝛼2𝐷𝑒𝑓𝑖𝑐𝑖𝑡𝑡+ 𝛼3𝑇𝑟𝑎𝑑𝑒𝑡+ 𝛼4(𝑇𝑟𝑎𝑑𝑒𝑡𝑥𝐷𝑒𝑓𝑖𝑐𝑖𝑡𝑡) + 𝛼5𝑙𝑛 ( 𝑖𝑡 𝛿𝑡) + 𝛼3ln(𝐻𝑢𝑚𝑎𝑛𝐶𝑎𝑝𝑖𝑡𝑎𝑙𝐼𝑛𝑑𝑒𝑥𝑡) + 𝛼3𝑡 + 𝜈𝑖 [1] ∆ ln(𝐺𝐷𝑃𝑡) = 𝛽0+ 𝛽1Δ𝑙𝑛 ( 𝑖𝑡 𝛿𝑡) + 𝛽2𝐷𝑒𝑓𝑖𝑐𝑖𝑡𝑡+ 𝜀𝑖 [2]
In the first regression, the steady state level of income per labor unit is the dependent variable and it is regressed on the previous year level of income (its lagged value), the variable of interest, Deficitt and other explanatory variables. These explanatory variables are used based on the Solow Growth model explained in the previous section. Furthermore, a variable will be used to represent the volume of trade openness in Portugal and an interaction
13
variable of trade and deficit will be created. Based on this interaction variable, it can be concluded whether fiscal consolidation has an influence in the trade openness and the resulting impact it has on the steady state level of income. To compare the skillfulness of the Portuguese workforce compared to the rest of the workforce, another variable is used, namely the human capital index. Additionally, a time-trend variable is added to the model.
The dependent variable of regression [2] is the difference of real gross domestic product (GDP) per labor hours between two consecutive periods. This is regressed on the difference of capital levels between two consecutive periods and government deficit. The natural logarithms of these variables are used for simplicity’s sake. The independent variable is government deficit and will be used to verify if it has a dragged effect on economic growth. Both dependent variables of regressions [1] and [2] are measured as being the real GDP over the average amount of hours worked per person engaged, giving therefore GDP per effective hour worked, which is in line with equation (6). The labor factor has been excluded from the right-side of the models as it is assumed that labor as constant returns to scale as stated in the literature and its already taken in the calculation of both dependent variables. Although ∆Ln(GDPt) seems to be the same variable in both regressions, they have different meanings and assumptions as they are explained by the nature of two different models.
4. Data
The data chosen to conduct this analysis is yearly-based, and regards the period between 1975 and 2014. The dataset was obtained from the Penn World tables 9.0 (PWT) (Feenstra et al., 2015) and Worldbank database. Since there are no data available for the direct variables of the Solow growth model, the variables that translate into the expanded version of the Solow Growth model were collected from the PWT database.
Initially, a relation of capital had to be made to get a variable that was capable of converting its effect into economic growth. Share of gross capital formation and average depreciation rate form the variable (it/δt), which will be used to estimate the amount of capital stock (K) and its changes through time.
From PWT data representing economic growth was collected. The dependent variable in both regression is calculated the same way. The most common instrument that has been investigated in past research, when assessing the impacts of austerity on the economy, is the growth of real GDP per capita (the nominal growth of GDP corrected for inflation). A slight variation is being used as the representation of economic growth in this analysis namely, growth of real GDP per labor hour as explained in section 3. Although GDP per labor was not explicitly available in PWT database, there existed data for both real GDP and the number of
14
average work hours of the Portuguese engaged population, so the simple equation below was performed to get to real GDP per capita:
Real GDP per capitat = Real GDPt (in mil. $) / av. Hours per person engagedt (in thousands.)
To be able to capture how efficient the Labor workforce is when compared to other countries, data of the Index of Human Capital was collected from PWT. This index is
calculated based on the years of schooling and the returns on education according to Feenstra et al. (2015) and will be added as a control variable in equation (3). For the fiscal
consolidations adjustments, the variable Deficitt is defined as the difference between
government tax revenues and its expenses resulting therefore as the primary balance. These two variables constitute the variable of interest that is going to be used as representation of the change in the fiscal stance of Portugal in the different periods from 1975 until 2014. Both variables are collected from the Worldbank database. Unfortunately, there is no information prior to 1975 about these variables. This coincides with the expectation, since Portugal had a dictatorship until 1974 and from the nature of such a regime, these data is either unreliable or untraceable which is the case in this situation. To measure the influence of trade openness on economic growth, data regarding exports and imports as a percentage of GDP from the Worldbank was retrieved. By adding these two variables up, the variable Openness is constructed.
5. Results and Findings
In this section, the results and findings will be discussed. Table 1 represents empirical results based on equation [1]. Table 2 represents the result analysis performed on regression [2]. Both regressions represent average levels of R2 (around 50%). Starting from the point in
which the models are correctly specified this should not be an alarming indication. Robust standard errors are used as advised by scholars particularly in this type of analysis as heteroskedastic standard errors might be present in both specifications. Furthermore, it is assumed that the necessary OLS assumptions hold.
5.1. Results and Findings from [1]
As stated in the previous section, some of the variables are taken in their log forms for easier interpretation. In this regression the impact of government deficit as measure of fiscal consolidation is assessed on the convergence theory explained in the work of Mankiw et. al. (1992). The first explanatory variable is significant which is consistent with the literature, since convergence is based on previous levels of Income per capita. This was done via using
15
its lagged value, which explains the extent to which is significant (p-value <0,001). The sign of this variable can be interpreted as if the previous year level of GDP per labor units increased by 1% then would affect the convergence path of the economy in negative since in order to reach to its steady state, then convergence of the present period would have to move in the inverse direction.
In this analysis Government deficit is defined as the difference between tax revenues and government expenses, which means that if fiscal consolidation would take place than a positive (the number would increase when compared to the absence of fiscal consolidation) change would take place in this variable. The sign is therefore not surprising, since it means that the more severe the fiscal consolidation adjustments adopted by a Government, the more the negative effect it has on the convergence path. This is logical since then a higher joint effort must be achieve to go to the steady state level of the economy. What does not go in line with the expectations formulated by analyzing previous literature is the fact that this effect is not statistically significant. This can possibly be explained by the limitations of this research which will be discussed in the next section.
The interaction variable between Trade and Fiscal consolidation is also not statistically different from 0, although its positive sign can be interpreted by economic theory. If austerity takes place, then possibly the markets will perceive this as a “problem fixing” situation, which will boost foreign trade of goods and services. Furthermore, most of the data collected dates to a period where Portugal was subjected to a floating exchange regime, this meaning that whenever the country was facing economic difficulties, the currency, Escudo, would depreciate which would likely increase exports, which has historically been the major component of trade as opposed to imports. Hence, the positive sign.
The next variable, which measures physical capital as stated previously, has a significant impact on the the steady level of Income. This goes according to the convergence path theory and therefore, the expectations created. A higher percentage increase in the level of capital leads to an higher steady level of Income per effective labor. As both the dependent variable and the variable in question are measured by their natural logarithms, it can also be inferred that indeed capital shows decreasing returns to scale as mentioned in the Solow growth model.
Human capital, also as predicted in the analysis conducted by Mankiw, Romer and Weil (1992), has an effect on the steady state level of Income, and therefore significantly affects the converge path of economic growth. Here again we see a positive relation with the level of economic growth, and setting therefore a higher level of steady state income since the coefficient for this variable is positive.
16
Table 1. Impact of Fiscal Consolidation on the Steady State Level of Income
VARIABLES ∆ln(GDP/Lt) Ln(GDP/Labour)t-1 -0.497*** (0.132) Deficit -0.00696 (0.0129) Trade 0.00193 (0.00296) Trade x Deficit 0.000181 (0.000203) Ln(Share Capital/Depreciation) 0.143*** (0.0367)
Human Capital Index 0.392***
(0.114) t 0.004 (0.004) Constant 0.996*** (0.259) Observations 39 R-squared 0.533
Notes: The numbers in the parentheses are the robust standard errors of the coefficient estimates. *** significant at the 1% level; ** significant at the 5% level; *significant at the 10% level
5.2. Results and Findings from [2]
In this regression, a different interpretation of the dependent variable and the effects of the explanatory variables has to be given since a different variation of a Cobb-Douglas production function is used. Specifically, in this situation the dependent variable means the actual growth of the economy measured by GDP per average hours of Labor per person engaged between two consecutive periods. The growth of the economy should be explained mainly by the change of the physical level in two consecutive periods as well, as the change in labor, in this case, labor hours, is already absorbed by the change in the dependent variable. It can be inferred that, as expected, the difference of accumulation of capital positively affects the growth rate of the economy and that this effect is statistically relevant. This is also predicted partially by the analysis of Nordhaus (1992).
The effect of government deficit, the instrument again used as measurement for fiscal consolidation, still defined as in the previous regression, directly affects the growth. Although its sign might come as a surprise taking into account the work of Nordhaus (1992), it goes in line with the theory given by the proponents of the expansionary view of fiscal adjustments. This is because, if fiscal consolidation would have a “drag” effect on economic growth, the the sign should be negative. In this case, since its presents a positive significant effect, it
17
works as a “boost” for economic growth, and therefore it can interpreted as a “win” from the expansionary effect of austerity on growth rather than a contractionary situation predicted by the Keynesian view.
Table 2. The Direct Effect of Government Deficit on Economic Growth
VARIABLES ∆ln(GDP/Lt)
∆Ln(Share Capital /Depreciation) 0.157***
(0.0145) Deficit 0.00317*** (0.000537) Constant 0.0779*** (0.0108) Observations 39 R-squared 0.516
Notes: The numbers in the parentheses are the robust standard errors of the coefficient estimates. *** significant at the 1% level; ** significant at the 5% level; *significant at the 10% level
6. Conclusions and Limitations
In this section, first an overview of the limitations and suggestions for further research on this topic are given. Finally, conclusions will be drawn based on the analysis of the literature and analysis of the results obtained.
The dataset analyzed presents already a short-coming by itself as only data ranging from 1974 is available. For such an analysis it would be better to make broad conclusions and this would perhaps be possible with a larger data set. Although due to the political situation prior to 1974 in Portugal, estimations of Government expenses and tax revenues could be made so that a larger sample is analyzed. Another problem that one might encounter, is the theoretical assumptions made in order to model the regressions, as there is no consensus between economists as considering the Solow model as the optimal representation on growth, this could lead, in the eyes of others scholars, a misspecification of these equations.
For further research it is advised to include an extra variable into at least the second equation to infer if for example the political party who is forming government has a
significant impact on growth. From this, another line of thought could be explored. For example, to infer if episodes of fiscal consolidation have a relation with the political standpoint experienced in Portugal. This could either be used as an instrument or directly in the equation analysis, depending on its characteristics, via a interaction variable between political party and fiscal consolidation method. Furthermore, a deeper analysis on the economic situation being experienced at a given moment and its impact on the success of a fiscal adjustment could be investigated by considering the Portuguese case. Furthermore, the
18
impact that different adjustments of revenues and expenses have on the success of fiscal consolidation, and therefore positive economic growth can be investigated.
Overall, it can be concluded that fiscal consolidation most likely has no considerable effect on the convergence path towards the steady state of the economy, as described
previously, in Portugal between 1975 to 2014. This however, can be misleading since fiscal consolidation has proven to have a substantial effect on economic growth by itself. This indicates that a severe episode of adjustments could set the steady state level of the economy in another level which could influence its path towards convergence.
19 References:
Alesina, A., & Ardagna, S. (2010). Large changes in fiscal policy: taxes versus spending. Tax policy and the economy, 24(1), 35-68.
Alesina, A., & Perotti, R. (1995). Fiscal expansions and adjustments in OECD countries. Economic policy, 10(21), 205-248.
Cabral, R. (2013). Austerity measures in crisis countries—results and impact on mid-term development. Intereconomics, 48(1), 4-32.
Caldas, J. C. (2012). The consequences of austerity policies in Portugal. International Policy
Analysis, 1-5.
Cunha, J. C., & Braz, C. R. (2007). Public expenditure and fiscal consolidation in Portugal.
OECD Journal on Budgeting, 6(4), 103-121.
Dias, T. (2013). The true cost of austerity and inequality: Portugal case study.
Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer (2015), "The Next Generation of the Penn World Table" American Economic Review, 105(10), 3150-3182, available for download atwww.ggdc.net/pwt
Giavazzi, F., & Pagano, M. (1990). Can severe fiscal contractions be expansionary? Tales of two small European countries. NBER macroeconomics annual, 5, 75-111.
Gorjão, P. (2012). Portugal and the straitjacket of the European financial crisis. The
International Spectator, 47(4), 64-68.
Jadhav, A., Neelankavil, J. P., & Andrews, D. (2013). Determinants of GDP Growth and the Impact of Austerity. The Journal of Applied Business and Economics, 15(1), 15.
Jayadev, A., & Konczal, M. (2010). The boom not the slump: The right time for austerity.
Kleis, M., & Moessinger, M. D. (2016). The long-run effect of fiscal consolidation on economic growth: Evidence from quantitative case studies.
20
Mankiw, N. G., Romer, D., & Weil, D. N. (1992). A contribution to the empirics of economic growth. The quarterly journal of economics, 107(2), 407-437.
McDermott, C. J., & Wescott, R. F. (1996). An empirical analysis of fiscal adjustments. Staff
Papers, 43(4), 725-753.
Nordhaus, W. D., Stavins, R. N., & Weitzman, M. L. (1992). Lethal model 2: the limits to growth revisited. Brookings papers on economic activity, 1992(2), 1-59.
Pescatori, A., Leigh, D., & Guajardo, J. (2011). Expansionary Austerity New International Evidence.
Wachtmeister, E., & Ödeen, F. (2015). Is austerity an effective recovery measure for the Eurozone?
