Download growth and political ideology in a democracy

GROWTH AND POLITICAL IDEOLOGY IN A DEMOCRACY:
THE FRENCH EXPERIENCE, 1871 - 2009
Mickaël Melki
François Facchini
-Centre d'économie de la Sorbonne, Paris 1, FranceDRAFT- WORKING PAPER
Abstract. This paper examines the influence of political ideology on economic growth in a democracy such as the French one - from 1871 to 2009. Indeed, the divide between the left and rightwing ideologies
has always shown within the sphere of voters and governments and ultimately led to various policies and
economic outcomes. This paper presents evidence suggesting that rightwing governments have generally
been more favorable to economic growth - and proposes to explore the channels of transmission through
which political ideology affects economic performance.
1. Introduction
Economists have traditionally neglected the role of ideology in the economic performance of states. This
seems paradoxical, as economists were at one time quite ideologically active with regard to their preferred
policies and institutions for implementation. Working to overcome this self-contradiction, a burgeoning
trend in literature recently emerged, working to test the impact of political ideology on economic growth,
inequalities, economic freedom, (Bjornskov 20051, 20082, Bjornskov and Potrafke, 2010), globalisation,
deregulation (Potrafke, 20093, 20104), public spending (Tellier, 20065) and public policy in general (Imbeau
et al, 2001). This article is written in the wake of this recent econometric trend in literature.
Nevertheless, the difference from the abovementioned literature lies in the following three points. Firstly,
all the current empirical studies which cover the affect of ideology on institutions or macroeconomic
performances share a particular feature; they are cross-sectional. They compare the effects of ideology
either for a group of countries or for states, lands or provinces inside a country. They are systematically
said to suffer from the same pathology: "fairly short observation periods… or focus on a small group of
countries." (Potrafke, 2010, 3) The proposition presented in this work is different and rather innovative. It
examines the long-term effects of political ideology on a single country, using a time-series analysis, which
is automatically immune to both pathologies. The second point of novelty stems from the area examined:
France. Until now, no work of this kind has been done on France. Although some indexes exist to
measure its political ideology, such as Beck et al (2001) - who put forward a classification of the
governments' parties between 1975 and 2000 - no long-term study is available on France and more
generally, for the other countries. Hence, devising such an index will stand as the third contribution of our
study to the partisan literature. In this manner, this article paves the way for long time analyses based on
quantitative history, with ever more reliable and accurate sources - such as the works of Maddison (1995,
2005), Barro and Ursua (2008)6 and the later-mentioned French academics.
Bjornskov, C. (2005). “Does Ideology Affect Economic Growth?” Public Choice, 123: 133 – 146.
Bjornskov, C. (2008). « The Growth-Inequality Association: Government Ideology Matters », Journal of Development
Economics, 87, 300 – 308.
3 Potrafke, N. (2009). “Did Globalization Restrict Partisan Politics? An Empirical Evaluation of Social Expenditures
in a Panel of OECD Countries”, Public Choice, 140, 105 – 124.
4 Potrafke, N. (2010). “Does Government Ideology Influence Deregulation of Product Markets? Empirical Evidence
from OECD Countries,” Public Choice, 143 (1-2), 135 – 155.
5 Tellier, G. (2006). “Public Expenditures in Canadian Provinces: An Empirical Study of Politico-Economic
Interactions,” Public Choice, vol.126, n°3/4 (March), 367 – 385.
6
Barro R. and Ursua J. (2008). Macroecomomic Cirses since 1870. Brookimgs Papers on Economic Activity.
1
2
1
Aside from these stakes, this article asks a simple question. Does political ideology affect economic
growth? Have rightwing governments been more or less favourable to growth than their counterparts on
the left? In other words, are the results found by Bjornskov (2005) for 25 OECD countries between 1975
and 2000, confirmed in the case of France? This finding may be opposed to the political business cycle
literature and the underlying empirical evidence - such as the works of Alesina (1987, 19887) - who noticed
the opposite effect for the United-States. In the U.S., Democrats and Republicans record similar growth
rates during the second part of their terms, but the former performed better during the early years of the
term. Thus Alesina (1988) agrees with Hibbs (1987, 19778), but minimises the partisan effect by finding
that it is limited to the first year of the term (Alesina, 1988, 34). At last, concerning our area, Dubois
(2005) used an Alesina-type model to estimate the effect of political ideology on the quarterly growth rate
in France from 1978 to 2005. But what about the examination of a longer period?
This paper is structured as follows. In the second section, the used growth model is specified and lists the
potential transmission mechanisms between ideology and growth (2). In the third section, it presents the
data: the ideology index, the French growth from 1871 to the present and a set of macroeconomic
variables (3). The forth section presents the tests and proposes to interpret the results (4). The fifth
section presents a conclusion (5) and the last one, the appendix, provides additional information (6).
2. Theoretical consideration
This section intends to explore the manner in which political ideology influences the economic
performance of a nation (GDP growth) (1.1), to specify then the nature of the effects of a right- or
leftwing ideology on the national economic performances (1.2) and to put forward the baseline model
used for the tests (1.3).
2.1 How to theorise the relationship between political ideology and economic growth. “Voters’
norms and perceptions of society might via their political ideology, affect economic performance by
affecting the policies chosen by politicians”(Bjornskov, 2005, 133). This causal relationship (Figure 1), the
key point in partisan literature, assumes on the one hand that the government and its bureaucracy
rigorously apply the choices of the sovereign voters (arrow 1) and on the other hand, that economic
policies do affect the economic performances of a country (arrow 2). The present article admits both
effects but does not dismiss the fact that governments can have their own objective functions (Mueller,
2003, 487-488)9 as well as the stability of institutions, the rules of the game. The political actors behave
both according to the rules and outside of them. Therefore voting, as a public choice process, leads to
slow changes in the rules of the game. It is part of an institutional path dependency process that accounts
for the diversity of national economic paths.
Figure 1: Transmission mechanism between ideology – vote and growth
VOTE
GOVERNMENT
1
3
ECONOMIC
POLICY
2
PERFORMANCE
2.2 Left – right divide and economic growth. The reason for which political ideology reflects the
existence of different norms and beliefs among voters.
In spite of the numerous debates in economics, the left-right divide remains a touchstone. In France, the
left-right opposition is acknowledged to be a touchstone to interpret the political spectrum (Jaffré et
Alesina, A. (1988). « Macroeconomic Policy and Politics, NBER Macroeconomics Annual, MIT Press, Cambridge,
MA 13 – 52.
Alesina, A. (1987). Macroeconomic Policy in a Two-party System as a Repeated game, Quarterly Journal of Economics
102 (3), 631 – 678.
8 Hibbs, D.A. (1977). Political Parties and Macroeconomic Policy, American Political Science Review, 71, (4), 1467 –
1487.
9
Mueller, D.C. (2003), Public Choice III, Cambridge, Cambridge University Press
7
2
Muxel, 1997, 71, 68, 95 10, Boy, Jaffré et Muxel, 199711, Mayer et al. 200712). The bulk of voters continue
to position themselves along a left-right axis (Mayer, 1997, 15). This behaviour is not arbitrary; but rather
lies on a real ideological consistency (Michelat et Tiberi, 200713).
It is possible to characterise the ideological spectrum more precisely. Grunberg and Schweisguth (1997,
175)14 propose to classify the French electorate according to three main features: (1) anti-universalism,
economic liberalism and religious orientation, (2) cultural and moral liberalism (anti-universalism, sexual
permissiveness and attitude towards rules) and (3) the existence of particular variables (for instance private
schools). Economic anti-liberalism is undoubtedly the most separating and important ideological
dimension to account for the left-right opposition and the underlying effects on the economic
performances of a country such as France. If the voter is rather in favour of a free-market economy, in
other words a small size of government, free trade, low regulation, low inflation, and the protection of
property rights; he or she is located on the right on the political spectrum. If he or she prefers a strong
government, able to protect the frontiers, to socialize a part of the domestic produce in the form of public
spending and for whom unemployment matters more than inflation; he or she will be of left-wing
ideology.
Taking a leaf out of the works of Budge et al. (2001, 22)15, Markussen (2008, 342)16 lists thirteen issues of
opposition between left and right-wing parties (Figure 2). They are political (majority democracy versus
constitutional democracy), strategic (war versus peace, colonization versus decolonization) and economic
(free-market economy versus planned economy, free-trade versus protectionism, cut in public spending versus
increase in public services). These main issues match the debates present in the French electorate, and in a
way, the American political spectrum - as theorized by Alesina (1988). The latter finds that democrat
governments are more favourable to economic growth than republican ones in the early years of their
terms, as they implement monetary (higher inflation) and expansionist budget (higher deficit and
spending) policies that are not immediately anticipated by voters (Alesina, 1988, 20). The ideological
divide is here between Keynesianism and monetarism. He concurs with the first findings of Hibbs (1977)
for the United-States on a standard theoretical macroeconomic basis with rational anticipation. However,
this presentation of the left-right divide is not relevant in the long run, and particularly so for the case of
France. Though, it is obviously difficult to consider the opposition between Keynesianism and
monetarism prior to the 1930's.
Figure 2: The 13 issues of the left-right divide
Jaffré, J. et Muxel, A. (1997). « Les repères politiques », dans Boy, D. et Mayer, N. (éditeur). L’électeur à ses raisons,
Presses de Sciences PO, Paris, 67 - 100.
11 Boy, D., Jaffré, J. et Muxel, A. (1997). « La dimension gauche – droite », dans Boy, D. et Mayer, N. (éditeur).
L’électeur à ses raisons, Presses de Sciences PO, Paris, 263 - 284.
12 Même si l’on souligne parfois l’affaiblissement du clivage gauche – droite ou le déplacement idéologiques au cours
de la campagne de 2007 les perceptions des enjeux sociaux et économiques restent structurés par le clivage gauche –
droite (Mayer, Palier, Le Hay, 2007, Le panel électoral français 2007. 1° vague – 29 mars – 21 avril 2007, CEVIPOF,
p.3). 18% des personnes se positionnant à gauche disent que « le chômeur peut trouver du travail s’il le veut
vraiment », 34% des électeurs de droite le pense.
13 Michelat, G. et Tibj, V. (2007). « Gauche, Centre, Droite et Vote. Permanence et mutation d’une opposition »,
Revue Française de Science Politique, 57, 3-4, 371 – 392.
14 Grunberg, G. et Schweisguth, E. (1997). «Recomposition idéologique », dans dans Boy, D. et Mayer, N. (éditeur).
L’électeur à ses raisons, Presses de Sciences PO, Paris, 139 - 178.
15 Budge, I., Klingenmann, H.-D., Volkens, A., Tannenbaum, E., Bara, J. (2001). Mapping policy preferences:
Estimates for parties, electors and governments 1945 – 1998, Oxford University Press, Oxford.
16 Markussen, S. (2008). « How the left prospers from prosperity », European Journal of Political Economy, 24, 329 – 342.
10
3
Note: extracted from Markussen, S. (2008). « How the Left Prospers from Prosperity », European Journal of Political
Economy, 24, 329 – 342.
Bjornskov (2005, 133) 17 proposes a similar prediction based on a different theory; that right-wing voters
are more inclined to believe in merit and a proper work ethic - since they are of the belief that inequalities
result not from hazard but from ones particular merit. In that sense, they have a more positive attitude
towards work and wealth creation. On the one hand, this stance can comprise to a certain extent the
disinterest of those who think they do not master their future and that only a large-scale institutional
change – the rise of socialism – can improve their living conditions. On the other hand it can, via the
voting process, drive political leaders to implement policies favourable to wealth creation. Thus, one can
expect right-wing governments to be more favourable to the free-market economy and therefore to
perform better than left-wing governments. Far from being partisan and biased, this assumption merely
matches the major findings of contemporary economic theory and the theoretical content of the left-right
divide on economic issues. It does not deal with the ideological oppositions related to mores (feminism,
homosexuality, permissiveness and so on), international issues (war or peace), the attitudes towards order
and authority. Rather, it focuses only on the effects of left-wing voters’ reluctance towards a free-market
economy and the economic outcome of leftwing governments - which share the same hostility.
2.3 Theorising French growth by introducing an ideology variable. Our aim is to introduce an
ideology variable in our explanations of long-term economic growth. Current literature already provides a
degree of explicative variables. Sala-I-Martin et al (2003)18 lists the 67 most common variables in the
literature, which they rank according to a Bayesian method (Bayesian Averaging of Classical Estimates (BACE)
Approach). They use a statistical argument to classify the different explanations of growth, rather than
theoretical ones. It is interesting to note that Sala-I-Martin et. al. never explicitly refers to political
ideology. Prior to choosing among the various growth models, we cope with a practical difficulty related
to the availability of data for a period as long as that which is presented here. Consequently, we will focus
only on the growth models which includes an ideological variable in their explanations. This significantly
narrows the scope of our study.
The political business cycle theory provides us with simple growth models. One such typical work of
literature is that of Alesina (1987, 1988). It holds the advantage of having been applied by Dubois (2005)19
for the period of 1978-2005 in France. In this model, growth is merely determined by world growth and
the political ideology (left/right) of the ruling government. However it’s drawback lies in its
underestimation of the role of ideology in political choices. This is because it represents a rational
anticipation model in which economic policies, inspired by a monetarist or Keynesian ideology, produce
effects only at the beginning of the political term. It is as though ideology held only an evanescent role.
« In particular, features such as merit assumptions differ according to ideology with rightwing individuals being more inclined to believe
that effort and ability pays off while leftwing individuals focus on the role of chance » (Bjornskov, 2005, 133).
18 Sala-I-Martin, X., Doppelhofer, G. and Miller, R.I. (2003). « Determinants of Long-Term Growth: A Bayesian
Averaging of Classical Estimates (BACE) Approach », American Economic Review, vol.94, n°1, 813 – 835.
19 Dubois, E. (2005). Economie politique et prévision conjoncturelle.Construction d’un modèle macroéconométrique avec prise en compte
des facteurs politiques. Thèse pour le Doctorat, Université de Paris 1 Panthéon – Sorbonne.
17
4
On the other hand Bjornskov (2005, 137) - who uses a standard, linear Barro-type model (Barro, 1997)20 proposes a more comprehensive and realistic model. It is more comprehensive because it takes into
account variables known to have a significant impact on national growth differentials, such as the initial
GDP per capita, the openness of an economy, human capital, the size of the government, the
government's share in GDP and the quality of institutions. It is more realistic because it accounts for the
inertia of voter ideology and the political culture of a country by dismissing the rational anticipation
assumption and thus assuming that policy effects are long-lasting.
On these grounds, our growth model is mainly inspired by the Bjornskov (2005), and sets aside the
rational anticipation assumption made by Alesina (1988). But, as our study focuses on a single country, as
does Alesina (1988), the initial GDP used in Bornskov (2005) in not more relevant in our case. It is not
necessary to check the convergence of the growth rates for a single country as opposed to a crosssectional analysis. We will replace this variable with another one, the annual European growth, which
represents the world growth variable in Alesina’s model. Therefore in our model, French growth is
explained by the following equation (E.1) which includes variables unanimously considered as the basic
determinants of growth (Baro, 1991): the European Growth (GrowthEU), the degree of openness of the
economy (Openness), human capital (Schooling), the intensity of left-wing ideology (Ideo) and the government
size, otherwise know as the share of public spending in the GDP (Government).
GrowthFRt   1   2GrowthEUt   3Schooling t   4Openess   5Governmentt   6 Ideot
()
()
()
(-)
(-)
3. Data
To examine the effect of political ideology on the economic growth of France, we gathered several pieces
of data coming from the quantitative economic history works carried out with France as its subject. The
list of variables (definition and sources) is provided in appendix, table A.1.
Initially, we built two kinds of variables to measure political ideology: a discrete and a continuous variable.
They are based on the composition of the French National Assembly, which is the Lower Chamber of the
French parliament, as opposed to the Senate, the Upper Chamber. A discrete variable (Government ideology
dummy) is the dummy variable which shows the value 0 when the majority of the Members of Parliament
(MP) is right-wing and 0 when it is left-wing. The continuous variable (Government ideology) indicates the
percentage of the left-wing seats. We did not use the Beck et all. (2001)21’s method, taken up by Bjornskov
(2005), which ranks the French parties along a left-right axis, and therefore allows for the distinction of an
extreme party from a moderate one.
The reason for our methodological choice being the emphasis our study would like to place on the
political ideology of the majority of the Assembly, all the more so because France was often governed by
left-of-centre or right-of-centre coalitions. Hence, it would not be relevant to organize the parties
according to their ideological intensity to the extent that the centrist parties, once they joined the majority,
tended to participate in right or left-wing governments. Concerning the calculation of our index (sources
described in appendix, table A.1), we did not include the few independent MP’s and worked to ensure that
their presence would not influence the political ideology of the majority. Furthermore, concerning election
years, during which the majority can shift ideology after the elections, we decided as a convention to
consider the composition of the Assembly after the elections22. Lastly, we would like to specify our
classification of the Vichy regime (1939-1944). We decided to attribute this as much to the right as to the
20
Baro R. (1997). Economic Growth in a cross-scetion of countries. Quaterly Journal of Economics, 106(2),
407-443.
21
Beck, T. Clarke, G., Groff A., Keefer P. and Walsh P. (2001). New tools in comparative political economy:
The database of political institutions. World Bank Economic Review, 15(1), 165-176.
For the estimation of our model, we check that this convention does not influence the results by testing
also the ideology variable that takes the values of the composition of the parliament before the elections
for the election years.
22
5
left. This drove us to assume that during these years, the Assembly was evenly shared between the two
political ideologies, and therefore the variable Government Ideology is coded 0,50 from 1939 to 1944. But, as
it is not possible to do so with a dummy variable, the variable Government Ideology Dummy is not available for
this period.
Concerning the other variables, we resorted to several databases (see the detail in the appendix; table A.1).
The explained variable is the annual GDP growth rate in France for the period of 1871-2009. We chose
1871 since it is the creation year of the Third Republic, the First and the Second Republic being very short
and did not fit the criteria of a modern democracy. To the best of our knowledge, we chose Maddison
(2005) as the most reliable source from among the different long-term data available for the French GDP.
We propose to compare it with the data of J-C. Toutain (1997)23, available only from 1890 (figure 1,
appendix)24. The explicative variables are described in detail in the table A.1 (appendix). The variable
European Growth that mainly stems from Maddison (2005), corresponds to the annual GDP growth rate of
11 European countries, which represented the main trade partners of France. Openness designates the
degree of openness of the French economy and is given by the volume of trade divided by the GDP and
the data is supplied by Asselain and Blancheton (2005)25. From Carry (1999)26, we built the Schooling
variable, which corresponds to the total domestic education spending - a proxy to measure the annual
investment in human capital. And finally we built the variable Government share of GDP, which represents
the government size, in other words,, the government’s share of total GDP. This was done by connecting
several series (Levy-Leboyer, 1985, 1990; Andre et Delorme, 1978a, 1978b 27, 1983; Fontvielle, 1976 28).
Finally, the index of institutional quality proposed by the Fraser Institute is available only from 1971 so we
chose to put it aside.
Table 2. Data
Variable
French Growth
Mean
Std. Deviation
Observations
2.40
7.35
139
2.29
3.28
139
4.39
4.85
139
28.97
8.79
139
31.38
15.19
125
54.27
21.16
139
0.58
0.49
135
European Growth
Schooling (ln)
Openness
Government share of
GDP
Government ideology
Government ideology
dummy
Toutain, J.C. (1997). “Le produit intérieur brut de la France, 1789 – 1990“, Economies et Sociétés, Histoire
économique quantitative, Série HEQ, n°1, n°11, 5 – 136. Toutain, J.C. (1987). « Le produit intérieur Brut de la
France de 1789 – 1982 », Economies et Sociétés, Série Histoire économique quantitative, numéro 15, mai.
24
In the estimation, we create a dummy variable for the years in which there is a significant gap between the
Maddison’s et Toutain’s data to control this uncertainty.
25 Asselain, J.C. et Blancheton, B. (2005). « Dynamique de l’ouverture internationale. Paradoxes, enjeux, éléments
d’interprétation à partir du cas de la France », Economies et Sociétés, Série « histoire économique quantitative », AF,
n°32, 49 – 179.
26 Carry, A. (1999). « Le compte satellite rétrospectif de l’éducation en France (1820 – 1996) », Economies et Sociétés,
Série AF, n°25, février – mars.
27 André, C. et Delorme, R. (1978). « L’évolution séculaire des dépenses publiques en France », Annales, Histoire,
Sciences Sociales, 33° année, n°2, 255 – 278.
28 Fontvielle, L. (1976). « Evolution et croissance de l’Etat Français 1815 – 1969 », Economies et Sociétés, Série « histoire
économique quantitative », septembre – octobre-décembre.
23
6
Which this annual data, a concern arises from the fact that ideology is measured across the same period as
growth. That could lead us to the identification issue - that is to say whether both variables are
endogenously determined. But this issue seems to be a minor one in previous works written on the effect
of political ideology on growth. For instance, among the previously mentioned works, Alesina (1988) paid
no attention to this potential endogeneity, and no more than Dubois (2005) or Potrafke (2010)29 - and
Bjornskov (2005) merely resorts to a Hausman. The connected literature does not offer appropriate
instruments for ideology, except in Bjornskov and Potrafke (2010)30, who had instrumented their ideology
indices with lagged variables of government employment and union density. In order to tackle this
problem, we propose two solutions. The first one consists of using the lagged variable of Ideology
Government (t–1) to see its impact on French Growth (in t). And we also propose to use government
employment to instrument our ideology variable. Indeed, the left-wing governments have shown a
tendency to hire more civil servants without proof of a direct impact on national growth. (TESTS TO DO)
4. Results
4.1 Does political ideology indeed affect economic growth? Before any test, it appears that the
French economic growth was shown to be higher under the legislature with a right-wing majority during
the period of 1871 – 2009. In fact, the average growth rate under a right-wing majority at the National
Assembly is almost 4% as compared to 2.40% under a left-wing one. This first stylised fact provides us
with a first indication concerning the effect of political ideology on economic performance. But, as we
need to create a control for the other variables, by reasoning all things being equal, and we need to explore
the way in which political ideology influences growth. To solve this problem, we will proceed in two
stages. The first one consists of comparing with an OLS method the economic performance of the right
and left-wing governments. The second one proposes to emphasize the transmission channels through
which ideology works on economic growth.
Table 3 : Results31
1
European Growth
Schooling
Openness
Government share of GDP
0.672***
(6,23)
4.087***
(2.83)
0.435***
(4.28)
0.065
(0.77)
Government ideology
Government ideology dummy
2
3
4
5
1971-2009 French growth
0.653***
0.338***
0.320***
0.349***
(5.92)
(3.13)
(2.92)
(3.17)
4.054***
6.582***
6.422***
6.447***
(2.91)
(5.22)
(5.29)
(5.39)
0.418***
0.369***
0.368***
0.362***
(4.03)
(3.93)
(4.03)
(3.87)
0.016
(0.19)
-0.022**
-0.021*
(-1.97)
(-1.72)
-0.956**
(-1.99)
Government ideology lagged
Observations
Adjusted R-squared
F-statistic
6
0.310***
(2.90)
6.222***
(5.20)
0.368***
(4.24)
-0.021*
(-1.77)
125
0.712
22.897
125
0.736
22.637
139
0.860
71.713
139
0.862
67.791
139
0.863
68.049
139
0.863
67.670
Potrafke, N., (2010) “Forthcoming, Political Cycles and Economic Performance in OECD Countries:
Empirical Evidence from 1951-2006,” Public Choice.
30
Bjornskov C. and N. Potrafke, Forthcoming (2010) “Political Ideology and Economic Freedom across
Canadian Provinces” Eastern Economic Journal.
29
Schooling, Openness and Government share of GDP being the stationary variable of
ln(Schooling), Openness and Government share of GDP respectively.
31
7
Note: Absolute value of t statistic in brackets; * significant at 10 percent; ** significant at 5 percent; ***
significant at 1%.
For the first stage, we propose two forms of regression, the first-kind regressions (Table 3, columns 1, 2)
without the war years and including our variable Government share of GDP - which is not available during the
war period; whereas the second-kind regressions (Table 3, columns 3, 4, 5, 6) include the war years but
dismiss the variable related to the government size. Whatever the test, we find as expected, that the
European Growth, the human capital (total education spending) and the openness degree have a positive and
significant (at 1%) impact on the French growth. The Schooling variable has a particularly prominent
influence, which contrasts with Bjornskov (2005, 137, 139) - who finds that the Schooling variable is never
significant for his panel. This divergence can be explained by the double effect of education spending,
emphasized by the latest long-term studies on France. For instance, Diebolt and Trabelsi (2009) 32 show
that economic growth produces knowledge but the former is not always accounted for by the production
of knowledge. For the two other variables, the results comply with the literature, but do not escape from
the controversy on the openness degree as an appropriate index - and from the sense of the relationship
between trade openness and growth. Bairoch (1993, Chapter 13, 188) 33 considers that “economic growth
has been one of the driving forces behind free trade rather than the opposite.” Finally, we can notice in
the first-kind tests, that the government share of GDP is not significant. This result does not tally with
Bjornskov’s finding - but it is not disconcerting in view of the related literature on France. From
Fontvielle (1976), a consensus appeared to consider public spending as a governments’ reaction to
economic growth and depression. From the beginning of the 19th century to the Second World War,
public spending would be counter-cyclical and since then, has become pro-cyclical.
As for the ideology variable, we find out as expected, that it contains a significant and negative impact on
growth. That is to say, the left-wing majorities at the National Assembly have performed significantly
worse than the right-wing ones. To be more precise, according to the first-kind regression (column 2), we
can reasonably (5% significance) claim that a 10% increase in the number of leftwing seats in the law
chamber that is to say only 57 seats out of 577 under the 5th Republic34 reduces the French growth by
0.2%. If we consider the whole period of 1871-2009 by integrating the 14 missing observations into the
regression and therefore putting aside the Government share variable - column 4 shows that political ideology
works in the same proportion but becomes significant at the 10% level. We also used a lagged variable of
the political ideology to consider the delayed effects of policies on economic performance - and in parallel
- controlling for the potential endogeneity. Regardless, the results remain unchanged (column 6). Finally,
we use a dummy variable of the government ideology and the results speak volumes - since in more than
95% of the cases (5% significance), growth is higher by almost 1% under a right-wing majority than under
a left-wing majority. In view of this first battery of tests, we can reasonably conclude the first step of our
analysis – which holds that since the beginning of democracy in France, the left-wing parties in the French
National Assembly have performed noticeably worse than the right-wing groups.
4.2 How does political ideology affect economic growth?
5. Conclusion
TO DO
Diebolt, C. and Trabelsi, J. (2009). “Human Capital and french macreconomic growth in the long run," Economies et
Sociétés, “Histoire économique quantitative,” AF, n°40, 901 – 918. Carry, A. (1999). « Le compte satellite rétrospectif
de l’éducation en France (1820 – 1996) », Economies et Sociétés, Série AF, n°25, février – mars.
33
Bairoch, P. (1999). Mythes et paradoxes de l’histoire économique/ trad. Anglais, Anne Saint-Girons : potf.
Jean-Charles Asselain – Paris, La Découverte.
34
After 19??.
32
8
6. Appendix
Table A.1. Data Sources
1-French Growth
France annual GDP growth rate
Component 1: 1871-2008 Maddisson (2005): France GDP, million 1990 International Geary-Khamis dollars
Component 2: 2009
National accounts- INSEE (National Institute of Statistics and Economic Studies) : France GDP, million
2000 euros
2-European Growth
Europe annual GDP growth rate
Maddisson (2005): GDP Total 12 Western Europe (Austria, Belgium, Denmark, Finland, Germany, Italy,
Component 1: 1871-2008 Netherlands, Norway, Sweden, Switzerland, United-Kingdom), million 1990 International Geary-Khamis
dollars
Component 2: 2009
3-Schooling
EUROSTAT: European Union GDP, million 2000 euros
Total domestic education spending
Component 1: 1871-1996 Carry (1999): Total domestic education spending, million current euros
Component 2: 1984-2009
data provided by the Ministry of National Education, on the website: media.education.gouv.fr: Total
domestic education spending, million current euros
4-Openness
Degree of openness of the French economy, given by the sum of the importations and exportations as a
percentage of the GDP
Component 1: 1871-2002
Asselain and Blancheton (2005): goods importations as a percentage of the (current price) GDP, goods
exportations as a percentage of the (current price) GDP
Component 2: 1960-2009 World Bank: the volume of trade as a percentage of the GDP
5-Government share of
GDP
Government’s share of total GDP
Component 1: 1871-1913 Levy-Leboyer and Bourguignon (1985, 1990): government consumption
Component 2: 1869-1912
(discontinuous), 1920-1938, Andre and Delorme (1983): total public spending/ (constant price) GDP
1947-1974
Component 3: 1959-2009
National accounts- INSEE (National Institute of Statistics and Economic Studies ) : (Billion constant
euros) public administrations spending/(constant price) GDP
6-Government ideology
Percentage of the leftwing PM in the French National Assembly, French overseas departments et
territories excluded and without taking account of the non-party MP
Component 1: 1871-1941, Website of the French National Assembly: http://www.assemblee-nationale.fr/histoire/index.asp:
1946-2009
distribution of the seats in the French National Assembly according to the different political parties
9
Figure A.1 French GDP level (1869-2009)
French GDB - million 1990 International Geary-Khamis dollars
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
0
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Source: Madidison (2005).
Figure A.2 French GDP growth rate (1869-2009)
50.00%
French GDP growth rate
40.00%
30.00%
20.00%
10.00%
0.00%
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
-10.00%
-20.00%
Source: Maddison (2005).
10
Figure A.3 French GDP growth rate (1891-1985) according to two different sources
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
1890
-10.00%
1900
1910
1920
1930
1940
1950
1960
1970
1980
-20.00%
French Growth (CEPII)
-30.00%
French Growth (Maddison)
Source: Madidison (2005), CEPII database
Figure A.4 Western Europe and French GDP growth rate (1871-2009)
30.00%
25.00%
20.00%
15.00%
10.00%
5.00%
0.00%
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
-5.00%
-10.00%
-15.00%
Western Europe Growth
French Growth
Source: Maddison (2005).
11
Figure A.5 Western European and French GDP levels, million 1990 International Geary-Khamis dollars
(1869-2009)
6,000,000
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
1870
1890
1910
France GDP level
1930
1950
1970
1990
Western Europe GDP level
Source: Maddison (2005).
Figure A.6 Western Europe GDP levels, million 1990 International Geary-Khamis dollars (1869-2009)
6,000,000
Western Europe GDP level
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
1870
1890
1910
1930
1950
1970
1990
Source: Maddison (2005).
12
Figure A.7 Western Europe GDP growth rate (1871-2009)
10.00%
Western Europe Growth
5.00%
0.00%
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
-5.00%
-10.00%
-15.00%
Source: Maddison (2005).
Figure A.8 Ln (Total domestic education spending in France), million current euros (1871-2009)
12.00
LN (Total domestic education spending, France)
10.00
8.00
6.00
4.00
2.00
0.00
1870
1890
1910
1930
1950
1970
1990
-2.00
Source: Carry (1999), Ministry of National education
Figure A.9 Degree of openness of the French economy (1871-2009), percentage of the GDP
13
50
Degree of openess of the French economy (%)
45
40
35
30
25
20
15
10
5
1870
1890
1910
1930
1950
1970
1990
Source: Asselain and Blancheton (2005), World Bank
Figure A.10 Government’s share of total GDP in France (1871-2009), percentage of the GDP
55
Government share of GDP (%)
50
45
40
35
30
25
20
15
10
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Source: Levy-Leboyer and Bourguignon (1985, 1990), Andre and Delorme (1983), INSEE
14
Figure A.11 Percentage of the left-wing PM in the French National Assembly, Government ideology
100
Ideology
90
80
70
60
50
40
30
20
10
2007
2001
1995
1989
1983
1977
1971
1965
1959
1953
1947
1941
1935
1929
1923
1917
1911
1905
1899
1893
1887
1881
1875
1869
0
Source: Website of the French National Assembly: http://www.assemblee-nationale.fr/histoire/index.asp
15