Can Informal Institutions Reverse the Resource Curse?
Table of Content
1. Introduction
p. 3
2. Literature review
p. 5
3. Defining informal institutions
p. 8
4. Methodology and data collection
p. 10
5. Results and analysis
p. 12
5.1 Hofstede model
5.2 World Value Survey model
5.3 GLOBE project Model
6. Discussion
p. 28
7. Conclusion
p. 30
8. Limitations and implications
p. 30
9. References
p. 32
Appendix A
p. 34
Appendix B
p. 36
The objective of this paper is to investigate whether informal institutions have an effect on reversing the resource curse and, if this is case, if they could be a substitute for formal institutions. A regression analysis is done using three different datasets measuring informal institutions in terms of cultural dimensions. Outcomes show that when combined with informal institutions it is not always necessary to have high levels of institutional quality to reverse the resource curse. In fact the curse can even be reversed in some cases with low institutional quality by putting more emphasis on informal institutions. Also, this paper shows that sometimes it is not the level of institutional quality that matters, but the neglected negative effect of a cultural characteristic in society. Furthermore, the varying outcomes among the different measures of natural resources indicate that a customized approach is needed for each specific natural resource in order to improve natural resource management.
1. Introduction
One would expect countries rich in natural resources to have a strong economic development and growth. However, existing research has shown that resource richness of countries turns out to be a curse rather than a blessing. Sachs and Warner discovered in their empirical research (1970-1990) that economies with a high ratio of natural resource exports to GDP in 1970 tend to grow slower in the following period. Also, resource-poor economies to a great extent outperform resource-rich economies (Sachs and Warner, 1997a), this phenomenon is called the ‘resource curse’ (Gelb, 1988). More recent, research has shown that countries with an abundant amount of natural resources constitute both winners and losers. On average, countries rich in natural resources tend to lag behind in economic growth. Examples of this can be found when looking at resource-rich countries in Africa and Latin America such as Nigeria, Zambia and Venezuela and comparing them to resource-poor countries like Korea, Taiwan and Singapore, also known as the Asian Tigers. However, not every country with an abundant amount of resources is bound to be cursed, countries including Botswana, Australia and Norway are both rich in resources and have managed to have strong economic growth (Mehlum et al., 2006). Existing research has tried to find the underlying factor that distinguishes the ‘winners from the losers’ by looking at the problem from several perspectives, among which are a geographical, economic, social, and political perspective. In the literature review of this paper, each of these perspectives will be discussed more in-depth.
policy choices are a determining underlying factor in the difference between winners and losers within the natural resource curse. In other words, not so much the amount of natural resources, but more the extent to which a country is actually able to manage its natural resources determines whether a country can profit from its resource abundance or not. Taking a closer look at a country’s natural resource management policy and its determinants, history has shown that there is no such thing as ‘one efficient natural resources management policy that fits all countries’ (Gylfason, 2008; North, 1994). Also, Sachs and Warner (1997) state: ‘there must be something else going on beyond wasteful policies,’ indicating that there might be other non-economic factors than institutional quality stimulating the resource curse. In the past, research has been done on the relationship between religion and institutional quality, (La Porta et al., 1999) and on the relation between culture and economic performance (Guiso et al., 2006). More recently, Gorodnichenko and Roland (2010) published an article in which they empirically test the concept of culture and its relation to institutional quality and economic growth. However, the work of Gorodnichenko and Roland focuses only on the dimension of individualism vs. collectivism. Also, Minkov and Blagoev (2009) investigated the relation between culture and economic growth. They hereby specifically focus on determining a causal relation between the two concepts, which puts the reliability of the results at risk since many external factors influence economic growth.
2. Literature review
Existing literature on the resource curse can be divided in two main streams: the ‘explaining’ and ‘diagnosing’ literature streams. The aim of the explaining stream is to find possible explanations for the fact that some countries seem to benefit from their resource abundance while other countries do not. Hereby, researchers use different perspectives from several academic fields in order to find the underlying factors that distinguish winners from losers regarding the resource curse. Research in this literature stream is either purely theoretical or the potential underlying factor is tested on a multi-country sample of resource-rich countries. Whereas the explaining literature stream is based on inductive broad-orientated research, the diagnosing literature stream uses a more in-depth deductive approach. In this latter research stream, a resource-rich country (or a specific set of countries) is chosen and, based on a predefined set of symptoms of the resource curse, it is tested whether this country is suffering from the resource curse or not. Countries that have been popular to investigate in this literature stream are Nordic countries (Gylfason, 1999 and Larsen, 2004), Latin American countries (Sachs and Warner, 1999) Russia (Oomes and Kalcheva, 2007; Ollus and Barisitz, 2007; Borkó, 2007) Nigeria (Sala-i-Martin and Subramanian, 2003) and South East Europe (Égert, 2005). However, this paper falls under the ‘explaining’ category. Therefore the focus of the remaining part of the literature review will be on this research stream. Each of the perspectives used to find potential underlying factors to explain the resource curse will be discussed. The possible distinguishing factors that have been discussed in existing literature can be divided into four perspectives:
Geographical
countries, have managed to develop well. Therefore, potential determining factors depend not so much on the ‘destiny’ of a country, but should be sought more in the interference of people. Sachs and Warner (1997b) also note that although the geographical characteristics might not be the distinguishing factor, problems such as malaria should be addressed in order to enable countries to develop in the future.
Economic
Since the resource curse is directly connected to economic development, it is obvious to search for an economic explanation. Existing research has discussed possible economic explanations for the resource curse, among which are financial sector development (Saborowski, 2009), capital inflows and exchange rates (Lartey, 2008), learning process (Stijns, 2001) and the effect of providing international financial aid (Adam and Bevan, 2003). However, the most commonly discussed economic explanation in existing literature is the ‘Dutch Disease’. This concept originated in the 1970s when the Dutch gas exports caused an appreciation of the Guilder and thereby damaged the Dutch flower exports. (Iranzo, 2008). The Dutch Disease can be defined as a phenomenon whereby windfall revenues from natural resources cause real exchange rate appreciation, which in turn reduces the competitiveness of the manufacturing sector in that country (Oomes and Kalcheva, 2007). Resource-based exports are defined as agriculture, minerals, and fuels (Sachs and Warner, 1997). A boom in natural resources – for example oil - affects the national economy via two channels: (1) the resource movement effect: when oil prices increase, the demand for labor and capital increases in the oil sector causing the wages in this sector to increase as well. Assuming both labor and capital are mobile, both factors will move from the manufacturing to the oil sector, enabling oil productivity to increase. Meanwhile output and employment in the other sectors (manufacturing and services) will decline. (2) Spending effect: an increase in oil prices leads to higher wages and profits in the oil sector and, in this way, raises the demand in the economy. Part of this demand is for the domestic service sector, causing prices of services to rise as well. However, prices of oil and manufacturing goods are determined internationally and therefore not affected, causing a real exchange appreciation. Both effects lead to four consequences for the national economy which are labelled as symptoms of the Dutch Disease: (1) real exchange rate appreciation; (2) slower manufacturing growth; (3) faster service sector growth; and (4) higher overall wages (Algieri, 2006; Oomes and Kalcheva, 2007).
difference between winners and losers in the resource curse. In other words, it is not so much the amount of natural resources, but the extent to which a country is able to manage their natural resources that determines if a country can profit from its resource abundance (Ollus and Barisitz, 2007; Collier and Gunning, 1999; Mehlum et al., 2006; Boschini et al., 2007).
Political
Sachs and Warner concluded that: ‘the path of GDP in natural resource abundant economies would be
lower than it would have been in the same economies with optimal policies. But such economies would not necessarily grow slower than other resource-poor economies’ (Sachs and Warner, 1997a). On the
other hand, Knack and Keefer (1997) concluded that currently poor countries are not able to catch up with the wealthier countries unless they have a sustainable institutional environment. In this paper, institutional quality will be defined using the framework of Kaufmann et al. (2003) consisting of three pillars: (1) process by which a government is selected (voice and accountability) and replaced (political stability); (2) the government’s capacity to formulate (government effectiveness) and implement (regulatory quality) sound policies; (3) the respect of both citizens (rule of law) and the state (control of corruption) for the institutions that govern economic and social interactions. Mehlum et al. (2006) conducted further research on the relationship between the resource curse and institutional quality. Their results contrast the claim of Sachs and Warner that institutions do not play a role. According to Mehlum (et al.) countries rich in natural resources constitute both winners and losers when it comes to growth. What distinguishes a winner from a loser can be explained by differences in the quality of institutions. When institutional quality is low – ‘grabber friendly’ – natural resource abundance pushes aggregate income down, whereas when institutional quality is high – producer friendly – resource abundance increases income. They state that: ‘Institutions may be decisive for how
natural resources affect economic growth even if resource abundance has no effect on institutions. We claim that natural resources put the institutional arrangements to a test, so that the resource curse only appears in countries with inferior institutions.’
management policy that fits all countries’. Also, Sachs and Warner (1997a) state: ‘there must be
something else going on beyond wasteful policies’.
Social
Going beyond government policies and institutional quality requires us to take a closer look at the background on which government policies are based and institutional quality is developed. History plays a role influencing this background. Sachs and Warner touch upon this in their study on African economies (Sachs and Warner, 1997b). However, they also state that despite the influencing role, it is not a determining role when it comes to whether or not a country has stagnant economic development. Also, religion is mentioned in existing research as an influencing factor. While the early work of Weber (1958) on Protestantism and the spirit of capitalism and the more recent work by Landes (1998) on Catholic and Muslim religions and their relation to institutional development are foundational, they are mainly historically and theoretically based. Porta et al. (1999) have undertaken an empirical study investigating the relationship between religion and institutional quality and came to a similar conclusion, namely that overall, Protestant countries have better performing governments than Catholic and Muslim countries. A third influencing factor is the cultural background. At the turn of the century, the exploration of economic institutions became a popular topic in research, encouraging researchers to go beyond formal institutions and take a closer look at informal institutions, in particular culture (Guiso et al., 2006). Landes (1998) concluded in his book ‘The wealth and poverty of nations’ that when looking at the history of economic development, it shows that culture makes all the difference. In addition, more recent work by Tabellini (2007) concludes that culture is the missing link in how political and economic history has shaped how current institutions function.
3. Defining informal institutions
North (1994) defines institutions as ‘human devised constraints that structure human interaction,’ whereby one can distinguish between formal constraints (e.g. rules, laws, constitutions) and informal constraints (norms of behavior, self-imposed codes of conduct). Together with enforcement characteristics, they define the incentive structure of societies and specifically economies. Since North focuses more on norms of behavior and less on religion, his definition will be used as a working definition in this paper to operationalize the concept of informal institutions. The norms of behavior and self-imposed codes of conduct are reflected in a society’s culture. However, there is no standard definition of culture, since it constitutes a broad range of concepts including beliefs, values, history, preferences etc. Examples of definitions used in existing research culture are:
influence are two standard ones -beliefs (i.e., priors) and values (i.e., preferences)’ (Guiso et al., 2006)
- All over the world people are confronted with common challenges in life, such as having relationships with other people. Although these challenges are universal, each culture distinguishes itself from others by the specific solutions it chooses to these universal challenges (Trompenaars and Hampden-Turner, 1997)
Three datasets that measure the cultural norms and values in a society via cultural dimensions and are commonly used in existing literature are:
1) Hofstede dimensions: in the period 1967-1973 Hofstede analyzed the IBM employee values scores, which covered more than 60 countries. Based on this, he developed a framework that consists of five cultural dimensions: power distance, individualism, masculinity, uncertainty avoidance and long-term orientation (Hofstede, 2005). Gorodnicenko and Roland (2010) applied the Hofstede dimension by relating the individualism-collectivism dimension to economic growth.
2) GLOBE (Global Leadership and Organizational Behaviour Effectiveness) studies: a research program focused on culture and leadership. National cultures in 61 nations are examined along nine dimensions: performance orientation, future orientation, assertiveness, power distance, human orientation, institutional collectivism, in-group collectivism, uncertainty avoidance and gender egalitarianism. (House et al., 2002)
3) World Value Service (WVS): a worldwide network led by Ronald Inglehart studying changing values and their impact on social and political life. The study covers a sample of 97 societies and measure values concerning religion, gender roles, work motivations, democracy, good governance, social capital, political participation, and tolerance of other groups, environmental protection and subjective wellbeing (WVS Brochure, 2008). The WVS database has been used by Minkov and Blagoev (2009) in their work on the causal relation between culture and economic growth.
4. Methodology and data collection
In the empirical section of this study, I will discuss the general methodology and data collection. After this, I will discuss the implementation of the cultural dimension in the original model and the results and analysis for each cultural framework separately. Following the existing work of Sachs and Warner (1997a), Mehlum et al. (2006), Boschini et al. (2007) and Minkov and Blagoev (2009) a linear regression growth model will be used. Data regarding growth, natural resources and institutional quality will be extracted from the dataset developed by Boschini et al. (2007), therefore their empirical growth model will be used as a base:
growthi = Xiα + β1NRi + β2IQi + β3(NRi × IQi ) + εi (1)
Growth is measured as the average yearly growth rate of GDP. X is a vector of controls among which initial GDP per capita level, period averages of openness and investment ratios and dummy variables for sub-Saharan African and Latin American countries and a constant. The natural resource variable (denoted as NR) is measured in four ways: (a) the share of primary exports to GNP (NR1) (b) exports of ores and metals as a share of GDP (NR2) (c) the share of mineral production in GNP (NR3) (d) value of production of gold, silver and diamonds as a share of GDP (NR4). Furthermore, the variable IQ measures the concept of institutional quality, the interaction between the two variables is measured by NR x IQ and finally ε is an error term. In order to implement the variable of culture the empirical growth model can be extended in the following way:
growthi = Xiα + β1NRi + β2IQi + β3(NRi × IQi ) + β4CDi + β5(NRi x CDi) + εi (2)
The growth model is extended by CDi, which is the value of the cultural dimension extracted from the Hofstede model, WVS database or GLOBE study, and by NRixCDi1, expressing the interaction effect between natural resources and the cultural dimension. The regression is run separately for each combination of natural resources measurement and cultural dimension resulting in sixteen regressions for each model. Boschini et al. (2007) find that β1 is negative, which is in line with the classical resource curse literature stating natural resources have a negative effect on growth. In line with the idea that institutional quality has a positive effect on growth, β2 is found to be positive. Also, the interaction effect β3 is found to be positive for good enough levels of institutional quality.
1 Separate regressions were also run including the interaction effect CD*IQ. In all cases the addition of the third
All data regarding the original empirical growth model was provided by Boschini et al. (2007). This concerns data for the independent variables natural resources and institutional quality, whereas the former was explained above, the latter was developed by using (unweighted) averages of indexes for the quality of bureaucracy, corruption of government, rule of law, the risk of expropriation of private investment and repudiation of contracts by the government extracted from the earlier work of Keefer and Knack (2002). Also, the data on the dependent factor ‘growth’, as well as the control variables (level of GDP per capita in 1975, investment and trade openness) can be used from the existing dataset by Boschini et al. (2007) which were originally taken from the Penn World Tables mark 6.1.
5. Results and analysis
5.1 Hofstede model
Data and Hypotheses
Data for the Hofstede study was collected in 67 countries2. However, the database does not have a measure for the fifth dimension Long-term orientation for every country, since this research was only done on a sample of 23 countries. Since data availability for the long-term orientation dimension only applies to a few countries in the adjusted sample of this study, the dimension is excluded from this research. The remaining four cultural dimensions will be incorporated in the existing dataset from Boschini et al. (2007). In describing each of the cultural dimensions hereafter, the definitions as employed by Hofstede (2005) will be used. The first dimension is power distance, which refers to degree of expectation and acceptance of unequal power distribution within a society, as seen from below in the hierarchical structure. Todaro and Smith (2009) argue that an unequal distribution is often accompanied by unequal access to resources leading to economic inefficiency. Therefore, it is expected in this study that a high degree of power distance is negatively related to growth. The second dimension is individualism, which is the degree to which individuals are integrated into groups. In highly individualistic societies, ties between individuals are loose, that is, everyone needs to look after him/herself. Collectivistic societies are characterized by a high degree of loyalty. In this case, individuals belong to strong groups with close ties, in which they help and protect each other. Since strong social networks (as found in collectivistic societies) tend to have a positive effect on economic development (WDR ‘Attacking Poverty’ 2000/2001) it is more likely that individualism is negatively related to growth. Thirdly, masculinity is the distribution of roles between the genders. It is important to notice that this does not indicate the extent of female emancipation, rather masculinity refers to how the distribution of roles influences the specific solutions chosen to universal problems in a way that determines which values dominate the solution. Masculine values can be characterized as assertive and competitive, whereas feminine values are characterized as modest and caring. In this case, it is expected that masculinity - due to its assertive and competitive characteristics - is positively related to growth. The fourth dimension is uncertainty avoidance, which can be defined as the extent to which people in a certain culture tolerate uncertainty and ambiguity, and how comfortable they feel in new, unusual or unstructured situations. Societies that score high on the uncertainty avoidance dimension tend to minimize the risk of being exposed to such situations by strict laws and safety measures. Societies with low uncertainty avoidance aim to have as few rules as possible. Since a high degree of uncertainty avoidance tends to be accompanied by a more comprehensive set of informal rules, it is
2 In some exceptional cases Hofstede has taken an average measure for a group of similar countries. For example
more likely that uncertainty avoidance is positively related to growth. The expectations regarding the influence of the cultural dimension and the interaction effect NR*CD are also displayed in table 2.
The sample from Boschini ea. has 50 sample items overlapping with the Hofstede sample. The outcomes of the regressions for the adjusted sample are shown in Appendix A. Although correlations and coefficients slightly differ in terms of magnitude in both directions, the results of Boschini et al (2007) are relatively stable when adjusting the sample for the Hofstede sample size. The only significant differences appear in correlations with respect to Value production gold, silver, and diamonds as share of GDP (NR4). A possible explanation for this is the exclusion of two large producers: Botswana and Congo (Dem.). After adjusting for the sample size, the Hofstede data on the cultural dimensions is inserted in the database.
Results
The correlations for the original variables, as displayed in table 1, remain the same as in the original dataset. All cultural dimensions show a negative correlation to the dependent variable, average yearly growth rate of GDP, except for individualism. Another interesting aspect is the correlation among the cultural dimensions, for example, individualism is negatively related to power distance while uncertainty avoidance is positively related to power distance. However this would be a cultural study on its own and since it is not the main purpose of this research it will not be discussed in the remaining analysis. Table 2 displays the regression coefficients. Natural resources show a negative direction for all cultural dimensions except masculinity when using NR2 or NR4. The coefficient for institutional quality is positive in all cases and shows a strong significance. With respect to the newly added variables, power distance, masculinity and uncertainty avoidance show a positive direction, whereas individualism shows a negative direction except when using NR2. Also, only masculinity appears to be significant for NR2. The interaction effect between the cultural dimension and natural resources yields varying results. The interaction effect of power distance is positive while the interaction effect for individualism is negative except for NR4 and only the interaction effect of individualism for NR2 shows a strong significance of 1%. Masculinity and uncertainty avoidance both have a positive interaction effect, whereby the interaction effect of masculinity is significant for NR2.
Analysis
the negative direction of individualism are in line with the earlier stated expectations and the World Development Report (WDR ‘Attacking Poverty’ 2000/2001). The exception of the positive direction for individualism (found for NR2) can be explained by the fact that an indirect effect of collectivism is the lack of necessity to undertake economic incentive. In an individualistic society people are more expected to look after themselves and in that way undertake economic incentive which in its turn can stimulate economic growth. This explanation is also in line with outcomes of the study by Gorodnicenko and Roland (2010), which stated that a stronger degree of individualism in a society leads to more innovation due to the social rewards associated with innovation. Opposite to what was expected, power distance appears to be positively related to growth. In other words, a climate where unequal power distribution is more accepted is likely to see more positive economic growth. A possible explanation for this could be that a higher degree of acceptance is associated with a higher degree of political and social stability compared to a society where power distance is low and people are constantly trying to challenge and shift the distribution of power. Based on this line of reasoning, political and social stability ,and in this way power distance, can have a positive influence on economic growth. The derivative3 of the empirical model as explained in section 3.1 can be used to analyze the hypothesis regarding the influence of culture on the resource curse and if culture can actually reverse the resource curse. The combined effect of formal and informal institutions can reverse the resource curse in three possible ways:
- Type 1: Average to high levels for both institutional quality and the cultural dimension:
Power distance, individualism (NR3 and NR4), masculinity (NR3) and uncertainty avoidance (except NR1)
- Type 2: A high level of institutional quality and a low level of the cultural dimension:
Power distance (NR3) individualism (NR2 and NR4), masculinity (NR3) and uncertainty avoidance (NR4)
- Type 3:A high level of the cultural dimension and a low level of institutional quality:
Power distance (NR2 and NR4), individualism (NR4) and uncertainty avoidance (NR2)
It turns out that the resource curse can be reversed in case of average to high levels of both institutional quality and cultural dimension. This indicates that a high level of institutional quality is not a critical requirement since the resource curse can also be reversed at average levels of institutional quality. Moreover, the outcomes show that not only institutional quality is important, but also a sufficient level of informal institutions, since both are necessary to reverse the resource curse in case of type 1. Type 2 could indicate two things: (a) if a cultural dimension falls in both type 1 and type 2, this indicates that the resource curse can still be reversed although the cultural dimension is low and that in this case formal institutions are more important than informal institutions, for example with
3
masculinity and power distance (b) if the cultural dimension does not appear in type 1 and only in type 2 this could indicate that it is actually important to have a low level of that cultural dimension combined with a high level of institutional quality in order to reverse the resource curse, which is the case for individualism (NR2). Furthermore, individualism, power distance and uncertainty avoidance appear in type 2 and in type 3 showing varying results for different measures of natural resources. This could be explained by the needed level of innovativeness (individualism), regulations and safety measures (uncertainty avoidance), and social and political stability (power distance) for the production of a specific natural resource.
5.2 World Value Survey model
Data and hypotheses
The World Value Survey (WVS) has conducted research in five waves since 1981 for around 97 societies. In each wave, a standardized questionnaire was used consisting of more than 800 questions that measured values concerning perception of: life, environment, family, work, politics and society, religion and morale and national identity (WVS Brochure, 2008). In this research, four of these questions (items) will be incorporated in the existing dataset from Boschini et al. (2007). These items are intentionally selected based on relevance to the topic and availability. The first item is related to
trust and falls in the ‘Perception of life’ category. The specific question asked here is (question code
A165): ‘Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?’ Hereby two answers were possible: either ‘Most people can be trusted’ or ‘Can’t be too careful’. This item will be incorporated in the existing database by taking the percentage of people who responded with ‘Most people can be trusted’ to reflect the degree of trust in others in that society as measured by the WVS. Algan and Cohun (2007) state in their study on the relation between trust and economic development that trust could explain a substantial share of economic development in the post-war period, mainly via improving total factor productivity and the accumulation of human and physical capital. Thus, it is more likely that a higher degree of trust in others is positively related to growth and has a positive influence on the natural resource management.
The second item belongs to the attitudes towards society category and is concerned with income
equality. The question here asked is (question code: E035) ‘What is your opinion on the following
inequality is more likely to have a negative effect on economic growth since: (1) people on the lower side of the income distribution have limited access to credit which in turn creates less economic incentive and investment opportunity; (2) the rich elite does not save and invest proportion of their incomes to local economies; (3) people on the lower side of the income distribution often have poor living circumstances which lower their economic productivity and in turn lead to a slower growing economy; (4) in a highly unequal income society, the rich elite will spend most income on foreign luxury goods, whereas a more equal society with raising income levels for the poor can stimulate the demand for locally produced necessity products such as food and clothing. In this way, income inequality creates economic inefficiency. It is important to notice that this variable does not reflect the actual income inequality in a country, but ‘what is perceived to be best for society regarding income equality’. Based on this, a more favorable opinion towards income inequality - which is a higher score on the WVS item income equality - is expected to be negatively related to growth and have a negative effect on natural resource management.
The third item has to do with the attitude towards work. Although there are interesting questions in the WVS such as ‘does hard work lead to success’, the data available for these items is very limited. Therefore, this research will use the quality of hard work from the category ‘value qualities children are raised with at home’. The question asked regarding this item was (question code A030): ‘Here is a list of qualities that children can be encouraged to learn at home. Which, if any, do you consider to be especially important? Please choose up to five’. In this research, the percentage of the respondents who mentioned this quality will be taken as a measurement for ‘the importance of hard work as a quality children are raised with’. Blagoev and Minkov (2009) argue in their recent work that modern economic development is mainly a function of savings, investments and hard work. Therefore it is more likely in this study that a higher value attached to hard work is positively related to economic growth and has a positive influence on natural resource management. The last item extracted from the WVS database is concerned with what people expect from the government and what they believe is the responsibility of the government. The question asked for this item was (question code: E037): ‘What is your opinion on the following statement: People should take more responsibility to provide for themselves vs. the government should take more responsibility to ensure that everyone is provided for’. Respondents could answer on a ten-point scale where 1 corresponds to the opinion that people should take more responsibility and 10 to the opinion that the government should take more responsibility. In order to include this item in the existing database, the average score on the ten point scale is calculated, whereby a higher average score indicates a more favorable attitude towards
government responsibility. Assuming that a higher score of government responsibility indicates that -
discussed above in the WVS database. Despite small variations in magnitude, the main outcomes appear to remain stable when adjusting the sample to the WVS database as shown in appendix A.
Results
Correlations are shown in table 3. Although non-significant, all WVS items are positively correlated to the dependent variable growth. The correlations between the WVS items and institutional quality all appear to be significant, whereby trust is positively correlated to institutional quality and negatively to the rest of the cultural dimensions. An interesting outcome regarding the correlations among the cultural dimensions is that trust is significantly negatively related to the value of hard work and government responsibility. Table 4 displays the regression coefficients. It shows that the natural resource measures show a negative regression coefficient apart for income equality (NR2) and hard work (NR1, NR2, and NR4). On average, income equality tends to have a stronger magnitude whereas hard work tends to have a lower magnitude. In addition, only income equality (NR2) and trust (NR3) are significant. Institutional quality shows a strong consistency in its outcomes, whereby the direction is positive with a moderate magnitude and a significance of 1% in all cases. The interaction effect shows a positive direction when using NR3 and NR4, while the direction when using NR1 and NR2 varies and only government responsibility shows a consistent positive direction. Furthermore, income equality only shows significance when using NR3. Income equality, hard work and government responsibility all show a positive direction with the exception of hard work when using NR3. The opposite is true for trust which shows a negative direction except for NR2. Furthermore, none of the regression coefficients for the cultural dimensions are significant. Also, the interaction effect between natural resources and cultural dimensions appears to be positive for trust except for NR2 and for government responsibility. Hard work shows a negative direction except for NR3, whereas the outcomes for income equality vary with a positive direction for NR1 and NR2 and a negative direction for NR3 and NR4. Furthermore, only income equality for NR2 appears to be significant.
Analysis
Table 3: Correlations WVS Analysis
stronger tendency towards ‘can’t be too careful’ indicates that people are more risk-aware. In that case being more risk-aware can have a positive effect on the economic activities, because people try to protect and insure themselves against economic flaws. This could explain why a high score on trust is negatively related to economic growth.
The WVS items income equality and government responsibility were both expected to be negatively related to growth, but turn out to be positively related. The different outcome from what was expected could be caused by the ambiguity of how the question was phrased in the WVS study. Take, for example, a closer look at government responsibility, the exact question asked was: ‘What is your opinion on the following statement: People should take more responsibility to provide for themselves vs. the government should take more responsibility to ensure that everyone is provided for’. This question does not indicate if respondents should answer this question compared to their current situation or what they believe the responsibility of the government is in general. When setting expectations in this research, the current situation was taken into account, assuming that people expected more from their government than what they currently received. However, if the question is aimed to investigate government responsibility in general, meaning that a high score indicates people’s belief that the government should take main responsibility, this could actually indicate high institutional quality and would thus explain the positive relation between government responsibility and economic growth. Unfortunately, the word ‘more’ and also the word ‘larger’ for income equality in the question phrasing make the questions ambiguous. The possible combined effect of formal and informal institutions to reverse the resource curse can be categorized again using the same typology as was explained in the Hofstede model (p. 15):
- Type 1: Average to high levels for both institutional quality and the cultural dimension:
Trust (except NR2), income equality (NR3 and NR4), hard work (NR3) and government responsibility (except NR1)
- Type 2: A high level of institutional quality and a low level of the cultural dimension:
Trust (NR2), income equality (NR3) and hard work (NR3 and NR4)
- Type 3:A high level of the cultural dimension and despite a low level of institutional quality:
Trust (NR1 and NR4)
results occurring both in type 2 and in type 3. This could be explained by the differences in risks associated with the production of specific resources and the degree of necessary risk-awareness.
5.3 GLOBE project model
Data and hypotheses
The Global Leadership and Organizational Behaviour Effectiveness (GLOBE) project is a research program focusing on culture and leadership. Data for the GLOBE project was collected at the end of the 20th century (1993-2003) for a sample of 61 countries4 via a 7-point-Likert scale survey among over a thousand middle managers in food processing, finance and telecommunication industries. The development of the cultural dimensions in the GLOBE project was for a large part based on the Hofstede dimensions, which led to more recent research studies on the similarities and differences between GLOBE and Hofstede (Dorfman et al., 2006). Since this is not the objective of this paper, the identical Hofstede dimensions and the non-relevant cultural dimensions are not included in this paper. This results in four remaining GLOBE dimensions, which will be discussed below based on the definitions used by Dorfman et al. (2002). The first dimension is institutional collectivism (officially labeled as Collectivism I: societal collectivism), which reflects the extent to which societal, institutional and organizational practices stimulate and reward collective distribution of resources and collective action. Hereby, one would expect that a higher degree of institutional collectivism gives more people access to resources and encourages economic incentive, therefore institutional collectivism is expected to be positively related to economic growth. The second dimension is
performance orientation, which is the extent to which society or an organization encourages and
rewards people for performance improvement and excellence. The improvement of performance can be translated into the degree people are motivated to improve the efficiency and effectiveness of their work and, in this way, the stimulation of innovativeness. An innovative climate tends to stimulate economic performance, and thus economic growth. Therefore performance orientation and economic growth are expected to be positively related.
The third dimension is gender egalitarianism. One needs to be aware that this is a different dimension than Hofstede’s masculinity. Gender egalitarianism refers to the extent a society aims to minimize gender role differences and gender discrimination, whereas masculinity refers to which values dominate in decision-making but has nothing to do with the inequality between men and women. An opinion which is often advocated (World Development Reports by World Bank 2000/2001, 2003, 2008; Todaro and Smith, 2009), is that women play a critical role in economic development because:
4 In some exceptional cases the GLOBE study has taken separate measures for the same countries. For example
(1) women are in charge of the daily natural resource management like collecting food and water (2); women tend to spend the household money more efficiently than men by having food, clean water and health as their first priority; (3) investing in women is investing in the whole family and can significantly improve the level of health and education in a country. Hence, one of the eight Millennium Development Goals (United Nations, 2000) was to make progress towards gender equality and empowerment of women by eliminating gender disparities in primary and secondary education. Based on this, it is more likely that a higher degree of gender egalitarianism is positively related to economic growth. The fourth dimension is future orientation, which refers to the degree people in a society engage in future-orientated behavior, for example planning and investing in the future while delaying gratification. This dimension is based on the Hofstede framework, but, since the long-term orientation from Hofstede could not be used previously due to limited data availability, the GLOBE dimension will be in this study. Hereby it is expected that future orientation will be positively related to economic growth. The Boschini et al. (2007) sample has 42 countries overlapping with the GLOBE sample. The outcomes of regressing for the adjusted sample as displayed in Appendix A show that all regression coefficients are similar in direction, but with slightly different magnitudes, both increasing and decreasing. The significance appears to be stable whereby significant levels differ and some coefficients are not significant in the adjusted sample such as institutional quality except when using NR2. The adjustments of the sample cause correlations to both increase and decrease in magnitude, whereas the overall tendency is an increase in magnitude. Also, although the significance levels differ slightly within the adjusted sample the degree of significance remains stable. All in all, despite minor changes, the outcomes appear to be valid and stable when adjusting the sample to the GLOBE sample.
Results
institutional quality is positive in all cases, except for collectivism when using NR2, and is significant for NR1. The cultural dimensions show a positive direction for future orientation and gender egalitarianism, whereas institutional collectivism appears to be negative except for NR3. Performance orientation has more varying results whereby it shows a positive direction for NR1 and NR2 and a negative direction for NR3. None of them is significant. The interaction effect between natural resources and the cultural dimensions tends to be negative except for institutional collectivism and future orientation when using NR1.
Analysis
Despite minor differences in magnitude, the tendency of a positive direction for both institutional quality and the interaction effect between natural resources and institutional quality is in line with the original outcomes of Boschini et al. (2007) and the original sample adjusted to the WVS. Natural resources show both non-significant positive and negative directions in the WVS analysis, whereas it only appeared to be significantly negative in earlier work. Looking at the cultural dimensions, the positive direction for gender egalitarianism and future orientation is in line with the earlier stated expectations. In particular, the positive outcome for gender egalitarianism highlights that women can play a critical role in economic development, as advocated by several international organizations such as the World Bank. Despite this, it is not often acknowledged in developing countries. The positive direction for performance orientation when using NR1 and NR2 is also in line with the expectations. However, there is a negative direction for NR3. Since the direction varies among the different measures of natural resources, a possible explanation could be that it also depends on the degree of innovativeness needed for those specific resources. However, how this influences those specific sectors should be investigated in further research.
- Type 1: Average to high levels for both institutional quality and the cultural dimension:
Institutional collectivism (NR1) and future orientation (NR1)
- Type 2: A high level of institutional quality and a low level of the cultural dimension:
Performance orientation (NR2)
Institutional collectivism and future orientation falling in type 1 show necessity of a combined effect from both formal and informal institutions in order to reverse the resource curse. The occurrence of performance orientation in only type 2 could be explained by the fact that performance orientation can cause rent-seeking behavior. In this way, the resource curse can only be reversed if the institutional quality is high and the cultural dimension is low. Thus rent-seeking behavior is being avoided. Furthermore, gender egalitarianism seems unable to reverse the curse in any way.
6. Discussion
The three cultural models are all consistent in their positive direction for institutional quality and their tendency for a positive direction for the interaction effect NR*IQ, whereas natural resources vary in their outcomes showing both positive and negative directions. None of the cultural dimensions appear to be statistically significant except for masculinity in the Hofstede model when using exports of ores and metals as a share of GDP (NR2). It is interesting to see that the cultural dimensions have low correlations among each other. This shows the importance of using different cultural dimensions, and that each cultural characteristic of society has a unique effect on growth. Also, the significant differences in outcomes between the different ways of measuring natural resources show that it matters which specific natural resource is being produced and that the resource curse and factors influencing the resource curse can have different effects among the types of natural resources, which is also in line with the main conclusions of the work by Boschini et al (2007).
opposite of the ability to reverse the resource. An example is seen in the WVS model analysis. This indicates that it is important to look specifically at the influence of the cultural dimension on natural resource management, and not only on its effect on growth and economic development in general.
The analyses of the interaction effects using the derivative leads to six interesting influences of culture and institutional quality:
1) Only a combined effect with high levels of both institutional quality and the cultural dimension can reverse the resource curse, this is the case for future orientation.
2) A combined effect with average levels of both institutional quality and the cultural dimension can reverse the resource curse. This is the case for power distance, individualism, masculinity, uncertainty avoidance, trust, income equality, hard work, government responsibility and institutional collectivism.
3) Even for low levels of institutional quality, high levels of the cultural dimension can reverse the curse. this is true for power distance, individualism, uncertainty avoidance and trust. 4) Low levels of the cultural dimension combined with high levels of institutional quality are
necessary for reversing the resource curse, this holds for individualism, trust, hard work, and performance orientation.
5) Any level of institutional quality can reverse the resource curse, as long as the cultural dimension is low in value, this is the case for trust.
6) The effect of the cultural dimension varies in direction among the different measures of natural resources. This could be explained by the importance of a related factor for the production of the specific natural resources such as innovativeness (individualism) and risk awareness (trust).
any level of institutional quality as long as the level of the cultural dimension ‘trust’ remains low might indicate the importance of risk awareness in natural resource management. Finally, the mixed outcomes for individualism and trust again indicate that a customized approach in order to improve natural resource management is needed for each specific natural resource.
7. Conclusion
The objective of this paper has been to investigate the influence of informal institutions - in the form of cultural characteristics - in society on the resource curse. Hereby the aim is to see if these cultural characteristics have an effect on reversing the resource curse and, if this is the case, if informal institutions can even be a substitute for formal institutions. Building on the work by Boschini et al. (2007) by extending their formula with cultural dimensions and the interaction effect between culture and natural resources, a regression analyses was done for three cultural models: Hofstede, WVS, and Globe whereby each model constituted of four cultural dimensions. The outcomes are in line with the work by Boschini ea, although natural resources show both positive and negative directions whereas in the paper by Boschini (et al.) only a negative direction was found. Overall, it shows that Hofstede and GLOBE are more in line with the expectations and also shows more positive effects of culture on growth compared to WVS. However, GLOBE shows the least ability to reverse resource curse. The different ways formal and informal institutions can influence a possible reversion of the resource curse lead to four main conclusions. First, it shows it is not always necessary to have high levels of institutional quality, since in most cases a combination of average levels of the cultural dimension and institutional quality can reverse the resource curse. Secondly, it turns out the resource curse can still be reversed, even when institutional quality is low, by using the cultural dimension via informal institutions instead. Furthermore, it shows that sometimes not so much the level of institutional quality matters, but the neglected negative effect of a cultural dimension in society is more important, as seen with risk-awareness. Finally, the varying outcomes among the different measure of natural resources indicate that a customized approach in order to improve natural resource management is needed for each specific natural resource.
8. Limitations and implications
be that their past Communist systems would have biased the outcomes (Minkov and Blagoev, 2009). If these former Communist countries would have been included, it could have had a possible effect on the outcomes of for example income equality, institutional collectivism and government responsibility, since they reflect important elements of Communism. Furthermore, the model does not take into account other economic sectors that could influence economic growth, such as the manufacturing and service sector. Also, the model does not encounter other factors that could influence both the dependent and independent factors such as civil wars, natural disasters and international trade agreements. These limitations show opportunities for future studies to improve research on the resource curse.
The outcomes of this study have three main implications for both academic research as well as international organizations. First, it turns out that the resource curse can still be reversed even when institutional quality is low by using the cultural dimension instead. Hereby, the implication is to further investigate how civil society can - via informal institutions - improve natural resource management and perhaps even reverse the resource curse. Secondly, it shows that sometimes not so much the level of institutional quality matters, but the negative effect of a cultural dimension in a society is more important. The fact that this is the case for trust could indicate the importance of risk awareness in natural resource management. The implication would be to further investigate the importance of both trust and risk awareness in natural resource management. Also, the varying outcomes among the different natural resource measures indicate that a customized approach in order to improve natural resource management is needed for each specific natural resource. Whereby, it is necessary to focus on the specific effect of that cultural aspect on natural resource management rather than the cultural effect on economic growth in general.
Finally, this paper yields two implications for new areas of research. An interesting field of future research would be to have economic development in terms of human development instead of economic growth as a dependent variable. In this way, one can measure to what extent natural resource management actual benefits a country in terms of sustainable economic development. A possible way to measure this is by using the Human Development Index (HDI)5, which constitutes not only the level of income, but also the level of education and health (life expectancy). Also, a new dimension in the Hofstede model named ‘Indulgence vs. Restraints’ will be published at the end of 2010/beginning 2011. This new cultural dimension could yield interesting outcomes regarding the effect of informal institutions on natural resource management.
5
9. References
Adam, C. and Bevan, D. (2003), ‘Aid, public expenditure and Dutch disease’ Working Paper, vol. 2003-02. Centre for the Study of African Economies, Oxford University
Algan, Y. and Cahuc, P. (2007), ‘Social Attitudes and Economic Development: An Epidemiological Approach’ IZA/CEPR Discussion Papers 2935, Institute for the Study of Labor (IZA)
Algieri, B. (2006), ‘The Dutch Disease: Lessons from Russia’ prepared for Workshop on the ‘Impact of Oil Booms’, University of Paris.
Borkó, T. (2007), ‘The suspicion of Dutch disease in Russia and the ability of the government to counteract’ Working paper no. 35, IGEC European Center.
Boschini, A. D., Pettersson, J. and Roine, J. (2007), ‘Resource Curse or Not: a Question of Appropriability’
Scandinavian Journal of Economics Vol 109, No 3, pp. 593-617
Collier, P. and Gunning J.W. (1999), ‘Why Has Africa Grown Slowly?’ Journal of Economic Perspectives
vol. 13, pp. 3–22
Dorfman, P. Hanges, P. House, R. and Javidan, M. (2002), ‘Understanding Cultures and Implicit Leadership Theories across the Globe: an Introduction to Project GLOBE’ Journal of World Business, Vol. 37, pp. 3-10
Dorfman, P.W. Hanges, P.J. House, R.J. Javidan, M. Sully de Luque, M. (2006), ‘Conceptualizing and measuring cultures and their consequences: a comparative review of GLOBE’s and Hofstede’s approaches’ Journal of International Business Studies Vol. 37, pp. 897-914
Égert, B. (2005), ‘Equilibrium Exchange Rates in South-Eastern Europe, Russia, Ukraine and Turkey: Healthy or (Dutch) Diseased?’ Economic Systems, vol. 29(2), 205-241.
Gelb, A. (1998), ‘Oil Windfalls: Blessing or Curse?’Oxford University Press, Oxford and New York
Gorodnichenko, Y. and Roland, G. 2010. ‘Culture, Institutions and the Wealth of Nations’. NBER working paper no. 16368, National Bureau of Economic Research, Cambridge.
Guiso, L. Sapienza, P. and Zingales, L. (2006), ‘Does Culture Affect Economic Outcomes?’
Journal of Economic Perspectives, vol 20, No. 2, pp. 23-48.
Gylfason, T. (1999), ‘Natural Resources and Economic Growth: A Nordic Perspective on the Dutch Disease ’was prepared for a UNU/WIDER research project on Resource Abundance and Economic Development.
Gylfason, T. (2001), ‘Lessons from the Dutch Disease: Causes, Treatment, and Cures’ prepared for STATOIL-ECON conference volume ‘The Paradox of Plenty’
Gylfason, T. (2008), ‘Oil management: Norway’s example’ African Growth Agenda.
Hofstede, G. (2005), ‘Cultures and Organizations: Software of the Mind’ McGraw-Hill Professional. Iranzo, S. (2008), ‘Delving Into Country Risk’. Documentos Ocasionales 0802. Madrid:
Banco de Espa. pp. 11-77
Kaufmann, D. Kraay, A. and Mastruzzi, M. (2003), ‘Governance Matters III: Governance Indicators for 1996–2002’ World Bank Policy Research Working Paper No. 3106, Washington, D.C.
Keefer, P. and Knack, S. (1997), ‘Why Don’t poor Countries Catch Up? A Cross-National Test of an Institutional Explanation’ Economic Inquiry, Vol. 35, pp. 590-602
Keefer, P. and Knack, S. (2002), ‘Polarization, Politics, and Property Rights: Links between Inequality and Growth’ Public Choice, vol. 111, pp. 127–154
La Porta, R. Lopez-de-Silanes, F. Shleifer, A. and Vishny, R. (1999), ‘The Quality of
Government’ Journal of Law, Economics and Organization vol.15, pp. 222-279, Oxford University Press
Larsen, E.R. (2004), ‘Escaping the Resource Curse and the Dutch Disease? When and Why Norway Caught up with and Forged ahead of Its Neighbors’ Statistics Norway, Research Department.
Lartey, E. (2008), ‘Capital Inflows, Dutch Disease Effects and Monetary Policy in a
Small Open Economy’, Review of International Economics vol. 16, No. 5, pp. 971–989.
Mehlum, H. Moene, K. and Torvik, R. (2006), ‘Institutions and The Resource Curse’ Economic Journal vol.
116, pp. 1 –20. Royal Economic Society.
Minkov, M. and Blagoev, V. (2009), ‘Cultural values predict subsequent economic growth’ International
Journal of Cross-Cultural Management vol. 9, No. 1, pp. 5-24
North, D.C. (1994), ‘Economic Performance through Time’ The American Economic Review, Vol. 84, No. 3 pp. 359-368
Ollus, S. and Barisitz, S. (2007), ‘The Russian Non-Fuel Sector: Signs of Dutch Disease?’ BOFIT Discussion Paper No 2.
Oomes, N. and Kalcheva, K. (2007) ‘Diagnosing Dutch Disease: Does Russia Have the Symptoms?’ BOFIT Discussion Paper No 6.
Paldam, M. (1997), ‘Dutch Disease and Rent Seeking: The Greenland Model’ European Journal of Political
Economy Vol. 13 (August), pp. 591-614
Palma, J. (2005), Four sources of ‘de-industrialisation’ and a new concept of the ‘Dutch disease’. Mimeo, Faculty of Economics and Politics, University of Cambridge, UK
Saborowski, C. (2009), ‘Capital Inflows and the Real Exchange Rate: Can Financial Development Cure the Dutch Disease?’ IMF working paper 09/20
Sachs, J. D. and Warner, A. M. (1997a), ‘Natural resource abundance and economic growth’ (revised Version, original version 1995) Working Paper, Harvard University.
Sachs, J.D. Warner, M. (1997b). ‘Sources of slow growth in African Economies’ Journal of African
Economies vol. 6, pp. 335-76
Sachs, J. D. and Warner, A. M. (1999), ‘The big push, natural resource booms and growth’ Journal of
Development Economics Vol. 59 pp. 43–76
Sala-i-Martin, X. and Subramanian, A. (2003), ‘Addressing the Natural Resource Curse: An Illustration from Nigeria.’ IMF Working Paper 03/019. International Monetary Fund, Washington, D.C.
Same, A. T. (2008), ‘Mineral-Rich Countries and Dutch Disease: Understanding the
Macroeconomic Implications of Windfalls and the Development Prospects: The Case of Equatoria’ Guinea’ The World Bank, African Region, Economic management Department.
Stijns, J. (2001), ‘Natural Resource Abundance and Economic Growth Revisited’ Mimeo, Department of Economics, University of California at Berkeley Tabellini, G. (2007), ‘Culture and Institutions’ IGIER working paper
Todaro, M.P. and Smith, S.C. (2009), ‘Economic Development’ Pearson, Addison-Wesley publishers, 10 th
edition
Trompenaars, F. and Hampden-Turner, C. (1997), ‘Riding the Waves of Culture (Understanding Cultural Diversity in Business)’ London: Nicolas Brearley Publisher
United Nations ‘Millennium Development Goals: 2015’ www.un.org/millenniumgoals (2000)
Weber. M. (1958), ‘The Protestant Ethic and the Spirit of Capitalism’ New York: Charles Scribner’s Sons World Development Report: Agriculture for Development (2008) provided by the World Bank, Washington
D.C.
World Development Report: Attacking Poverty (2000/2001) provided by the World Bank, Washington D.C. World Development Report: Sustainable Development in a Dynamic World (2003) provided by the World
Bank, Washington D.C.
