Can Dutch Disease be successfully treated? Lessons from the Netherlands’ experience for green political economy

John Cameron*

Abstract

The phrase ‘Dutch disease’ as a form of ‘resource curse’ occurs frequently in the literature on natural resources and development. It is usually associated with the transmission mechanism between the windfall effects linking exporting large amounts of a low unit cost, natural resource to the performance of the manufacturing sector. The mechanism focuses upon changes in exchange rates that threaten to shift the economy towards falling manufactured exports and increased consumption of manufactured imports at the expense of domestic manufacturing. The results of the comparison do suggest that after the discovery of off-shore natural gas in 1959 there was a period that can be characterised as distributional ‘Dutch Disease’. But changes in macro-economic, micro-economic, and human resource policies from the mid-1980s shifted the economy towards more productive use of gas revenues culminating in 1994 with the establishment of a ‘Fund for Enhancement of the Economic Structure’. The Netherlands’ governance regime proved capable of exercising agency to make this change. Looking towards green political economy, the results indicate the need for a hybrid institutional economics and political economy approach to understanding the impact of natural resource extraction and the importance of learning through evidence-based, societal deliberative learning processes informing an effective accountable governance regime.

*Corresponding author: Dr John Cameron, Emeritus Associate Professor; International Institute of Social Studies, Erasmus University of Rotterdam; The Netherlands.e-mail: cameron@iss.nlThanks go to Ayesha Mehboob whose Master’s Research Paper at the Institute of Social Studies, Erasmus University of Rotterdam provided material and inspiration for this paper. But John Cameron accepts full responsibility for this paper. John Cameron has been researching development issues for forty five years. He is a formal economist by original training, but has published journal articles claiming that relevant economics must always have a strong institutional awareness, including recent articles applying institutional economics to the political economy of environmental change.

‘In 1959 the joint [Shell and Exxon] company found natural gas off the coast of Holland [sic] at Groningen – a discovery which was to do more for the economy of Holland than the Dutch East Indies ever did.’ (Sampson, 1975: 192)

‘Petroleum was both a blessing and a curse to the Latin American nations in which it was found. The oil industry offered higher wages, provided an efficient fuel for transport and industry, and paid revenues that enabled the state to expand political, economic, and social programs. But, before 1930, petroleum development also meant that large foreign enterprises assumed de facto control of national resources ..’ (Brown, 1985: 362)

1. Introduction

The literature on the economic impact of natural resource extraction shows mixed

results. According to Van der Ploeg (2011), the empirical evidence suggests that, in analysing

natural resources as curse or blessing, one should expect either possibility. During the 19th

century and the first half of the 20th century, economies with relatively abundant natural

resources grew rapidly, such as, Australia, Scandinavia, and the United States of America

(Bravo-Ortega et. al. 2005). However, in the second half of the 20th century, many natural

resource abundant countries, experienced slower growth than less well endowed economies.

Latin American explanations for this phenomenon included Prebisch-Singer ‘dependency’

theory and Gunder Frank’s ‘underdevelopment’ theory.

The concept of ‘natural resource curse’ or ‘Dutch Disease’ describing a negative impact

on long-term growth due to excessive reliance on exporting an abundance of primary

commodities became widely accepted. The concept has been applied beyond natural

resources to include disproportional presence (a large ‘excess’ above local absorptive

capacity) of any commodity in a local economy with its own currency (Cameron and

Ndhlovu, 2012). For instance, Raghuram and Subramanian (2011) apply the concept to

international Official Development Assistance in heavily ‘aided’ economies finding a classic

Dutch Disease dampening effect on manufacturing exports.

The term ‘natural resource curse’ was coined by British economist Richard M. Auty in

1993. Auty (2001a, 2001b) explored the phenomenon of natural resource abundance in

various countries, for example, oil in Nigeria; diamonds in Sierra Leone and Angola; tin in

Bolivia. He suggested these economies grew more slowly as compared with natural resource

poor countries, for example, Hong Kong; Singapore; South Korea and Taiwan. Auty (2007a,

2007b) explained that support for resource curse grew when Sachs and Warner in 1995 used

regression analysis to compare economic performance during the time period 1970-1989.

Sachs and Warner (1995a, 1995b) found a statistically significant, negative relationship

between natural resource intensity and economic growth over those twenty years. Sachs and

Warner (1999) concluded that, though most resource rich governments introduced more

protective trade policies to counter the employment-diminishing effects of resource

abundance, these policies failed to remove the resource curse effect.

McKinley (2008) conducted a case study of a range of countries, including Angola and

Zambia in Sub-Saharan Africa; Brazil, Chile, Trinidad and Tobago, and Venezuela in Latin

America and the Caribbean; Yemen in the Middle East and Uzbekistan in Central Asia.

Initially, each of these countries enjoyed the benefits of an increase in revenue from the

export of primary commodities, such as oil, gold and copper. However, they experienced a

longer term decline in revenues due to falling global demand for primary commodities in the

period researched. McKinley claimed that the large flow of export revenues in these countries

have distorted their macroeconomic stability and slowed growth in their other economic

sectors, especially manufacturing. In the Latin American context, Sachs and Warner (1999)

conclude:

‘Empirically, resource booms seem to have done little to generate long-term growth, and

may in fact have hindered growth on average. The resource booms in Bolivia, Mexico

and Venezuela did not permanently raise the level of per-capita GDP, and were followed

by a growth slowdown rather than increase. The resource boom in Ecuador appears to

have raised the level of GDP initially but was not followed by faster growth.’ (Sachs and

Warner, 1999: 64)

This quote also introduces a time dimension and a sense of process that are important to

this paper.

Sachs and Warner (2001) provide a theoretical explanation of the Dutch Disease using an

endogenous growth model (see also Jerome, et.al., 2005). There are three sectors in the

model, a traded manufacturing sector, a non-traded service sector and a traded resource

sector. Endogenous growth arises because employment in the manufacturing sector generates

improvements in human capital, as a by-product. When natural resources are discovered, they

raise the level of income for a number of periods. Now, a part of this income is spent on the

non-traded service sector, drawing resources away from manufacturing, in turn, increased

demand for manufactures is satisfied through imports. Further, the reduction in

manufacturing employment reduces the rate of growth, due to the decrease in the

accumulation of human capital.

Ranis identified six factors that may harm the achievement of sustained economic

development due to the presence of abundant natural resources. First, resource abundant

countries may ignore the importance of human development due to excessive resource rents.

Second, resource rents may create rent-seeking activity instead of productive activity. Third,

resource abundant countries may adopt a policy of uncompetitive import substitution

industrialization. Fourth, there is unequal distribution in resource rents due to specific

interests of elite class or monopoly groups. Fifth, there is less economic growth in the

absence of export diversification because the prices of natural resource exports are more

volatile as compared with the prices of manufactured goods. Final, Dutch Disease effects

may seriously weaken the competitiveness of the non-mining tradable sectors (summarised in

Auty 2001a). Auty (2001a) criticised Ranis for not including factors such as the dominant

system of landholding, the type of political state and the choice of the developmental strategy

between natural resource endowment and economic performance.

Davis and Tilton (2005) suggested that mineral booms, such as in the Netherlands since

the 1960s, are in itself not necessarily a problem. So-called Dutch disease can actually allow

a country to benefit from its new found mineral wealth by encouraging resources to flow

from other sectors of the economy to the booming natural resource extraction sector, as

mentioned by Gylfason (2001a). If natural capital assets are converted into human or

physical capital for use in this booming sector; then they can promote economic growth. And,

if natural (capital) assets are consumed, then they can lower current levels of poverty. So, in

either case, natural resources can enhance economic development in terms of both economic

growth and poverty reduction. But clearly such positive results depend on both the factor use

productive characteristics of the natural resource extraction sector and the distributional

characteristics of the governance regime.

A critical perspective on the inevitable negative resource curse effects was also offered

by Murshed (2008). He examined three arguments relating to the phenomenon of resource

curse. First, resource curse is a recent (post-1970) proposition and, historically, abundance of

natural resource endowments are blessings rather than curse. Second, there are various

macroeconomic mechanisms producing the conversion of resource booms into resource

curse, such as, spending effect due to excessive consumption, and a relative price effect due

to real exchange rate appreciation. Third, the presence of large resource revenues may be the

cause of adverse political responses, for example, a pre-existing rentier state; unaccountable

institutional settings, and endemic corruption. Murshed and Serino (2011) concluded that:

‘The findings of this paper have policy implications for countries richly endowed with

natural resource based products. They are likely to fail to grow in the long-run if they do

not succeed in diversifying their economies and export structure: a process that can be

started by moving into natural resource processing.’ (Murshed and Serino, 2011: 159)

This suggests the resource curse challenge lies not in the existence of natural resources or

natural resource revenues, but in institutional management those resources and revenues. The

role of institutional contexts is central to this paper building on the writing of North and

Thomas (1973) reflecting on the impact on Spanish economic development of mineral

extraction from Latin America and Mahdavy’s (1979) ‘rentier state’. But as in much of the

debate surrounding resource curse the econometric quantitative evidence is ambiguous and

sensitive to:

ontological assumptions on the plasticity of institutions and agencies capable of

‘reforming’ those institutions;

indicators used to proxy the variables of resource availability (e.g. flows and

stocks) and institutional types and/or quality;

the time periods and countries included in complex disaggregated panel data sets;

an endogeneity challenge in attributing the direction of causality between natural

resource availability and institutional types/quality,

For instance, on ontological confidence:

‘Recent pilot studies by the World Bank have developed a methodology to assess the

political economy of natural resource governance along the extractive industries value

chain. Such an approach can be developed by domestic researchers, NGOs in oil rich

countries, or development partners in collaboration with oil producing governments.’

(Al-Kasim et al, 2013: 146)

On specification:

‘we have split the sample in two subsamples of equal size, according to the quality of

institutions... the indication of a resource curse only appears for countries with inferior

institutions ..’ (Mehlum et al, 2006: 1)

On disaggregation:

‘Our results suggest that countries rich in ores and metals are ... the ones with the

largest negative effects from the resource, but also that they are the ones where

institutional quality really makes a difference for the outcome.’ (Boschini et al, 2012: 31)

And on endogeneity:

‘there is no evidence that resource abundance negatively affects institutional quality,

contradicting the hypothesis of an indirect natural resource curse ... beneficial growth

effects seem to diminish as institutional quality improves, although they remain strongly

positive overall.’ (Brunnschweiler, 2008: 412)

This paper accepts the ambiguities in research based on high levels of quantification and

complex panel data sets and addresses the four challenges by:

Ontologically assuming that there is deep, historically long duration, institutional

structuration, but there is a possibility of learning and modifying institutional behaviour albeit

over decades;

On specification of indicators, focusing on extraction rate flows of a specific natural

resource and their derived revenues plus exploratory data analysis of quantified time series

that stimulate reflection on qualitative institutional behaviour;

On disaggregation, this paper is a case study of a specific natural resource in a single

country (the Netherlands) aiming at possible generalisation by creating a theoretical framing

that could be applicable elsewhere, including Latin American countries;

On endogeneity, this paper returns to its ontological assumption of long duration,

institutional structuration with a potential for learning and thus runs elements of causality in

both directions.

The remainder of this paper explores the institutional frontier between

distributional and productive use of natural resource windfalls for the case of the Netherlands,

the original ‘home’ of Dutch Disease. The Netherlands offers a relatively long time series of

reasonably trustworthy economic data. It also allows an opportunity to conduct a comparison

with the sub-set of OECD economies that can be claimed to have some similar characteristics

apart from a natural resource windfall; al with relatively reliable data. While a completely

controlled cross-sectional experiment is impossible, comparing the Netherlands’ experiences

with the wider OECD experiences does offer an element of control for the effects of changes

in the global economy on relatively high income economies with strong manufacturing

sectors. The concluding section analyses the Netherlands’ experience and offers reflections

that may be useful for research on ‘extractivism’ in Latin America.

To help understand the analysis, it is useful to think in terms of three epistemological

stances derived from basic economics theory (Cole et al, 1992). Broadly neo-classical

economics theory sees a market-driven tendency towards a ‘leisured’ rentier economy as

structurally inevitable and welfare acceptable as windfall natural resource revenues are

distributed across society. Institutional economics theory is concerned about the negative

social effects of this ‘rentier’ tendency and emphasises the possibility of using windfall

natural resource revenues to increase productive investment with appropriate policy

interventions by a ‘reformed’ state. Political economy theory also sees the dangers of a drift

towards a rentier economy with a focus on distribution, but is more pessimistic about off-

setting policy interventions to increase productive activities at the national level.

2. The concept of ‘Dutch Disease’ in discussions of the Netherlands’ experience

The discovery of the vast Slochteren natural gas field in Groningen province in the

Netherlands in 1959 was a major structural economic event (Hutchison, 1994 and Di John,

2007). Despite being targeted for sanctions in 1973 when ‘Holland ... alone among the

Europeans had been embargoed as punishment for her pro-Israel policies’ (Sampson, 1975:

275), the Netherlands received some protection from Shell (Sampson, 1975: 275) and was

able to grow economically in a period when other higher income economies were struggling.

Although the natural gas revenues increased overall national wealth and improved the

balance of payments, it was feared that some sectors, including the manufacturing sector

would decline due to international competitiveness and adversely affect whole regions and

major segments of the work force. Such fears that a natural resource based export boom

would have adverse economic effects underpinned the concept of 'Dutch Disease'.

The Dutch Disease literature emphasises how natural resource based exports boom can

have adverse effects on manufacturing (Corden, 1984; Sachs and Warner, 1995a; 1995b;

1999; 2001) due to appreciation of the national currency exchange rate against the currencies

of major importers of manufactures (Corden, 1984), unproductive consumption of resource

revenues through higher social contributions and a more general increase in government

spending (Wierts and Schotten, 2008; Van der Ploeg, 2011), and reduction of investment in

human capital development (Gylfason, 2001a). But Gylfason (2001b) claimed that from the

late 1960s onwards, Netherlands’ exports of goods and services have increased from less than

40 per cent of GDP to nearly 60 per cent, a high ratio by international standards for a country

with 16 million inhabitants. The composition of these exports will be explored later in this

paper.

Corden (1984) claimed to find such Dutch Disease adverse effects on Netherlands’

manufacturing with increasing natural gas extraction in the nineteen sixties, driven by

appreciation of the Netherlands guilder’s real exchange rate. Corden (1984, p. 359) argued

that Dutch Disease in the Netherlands was not only the adverse effect on manufacturing, but

also the use of booming sector revenues for social welfare payments, which were not

sustainable and were difficult to reduce politically.

Wierts and Schotten (2008) emphasised consumption of the gas revenues (e.g., higher

social welfare payments), in the 1970s, as crucial to ‘Dutch Disease’, a term that not only

referred to the pressure exerted by the higher gas revenues on the real exchange rate, but also

to the ‘derailment’ of the real economy and public finance. Wierts and Schotten (2008) point

to the risk that a sharp rise in revenues from natural resources may encourage politicians to

use the temporary income to raise public sector welfare expenditures to a level that is

unsustainable once the natural resources are exhausted (Gylfason and Zoega, 2002; Van der

Ploeg and Pelhekke, 2009). Wierts and Schotten (2008) description of the causes of ‘Dutch

Disease’ supports the distributional rentier hypothesis with some of the policy pessimism of

political economy theorising. In this context, Gylfason (2001b) suggests that natural resource

abundance, in an economy like that of the Netherlands, may reduce private and public

incentives to accumulate human capital due to high levels of non-wage income, including

social welfare payments not funded from general taxation.

Following this line of argument that the experiences of the Netherlands may be more

complex than a simple curse versus blessing dichotomy, the methodology used in this paper

allows exploration of sub-periods in the Netherlands’ experiences of windfall income from

gas as a natural resource across a number of dimensions. This sub-division permits the

analysis to go beyond a dichotomous either curse or blessing and explore the mix of rentier

and productive uses of windfall resources over time.

3. The changing role of natural gas in the Netherland’s economy

The industrialization process in the Netherlands as compared with other Western Europe

countries was relatively late. The period 1850-1870 can be seen as a run-up period of

technical progress and the decade 1880-1890 as a transitional phase towards industrialisation.

Economic development in that decade was based on manufacturing such as engineering, ship

building and vehicle and engine manufacturing alongside consumer goods industries such as

those producing quinine, ready-made clothing, margarine, biscuits and electric light bulbs.

In 1904, a systematic search for useful natural resources had started with the

establishment of the National Institute for the Prospecting of Minerals. The exploitation of

coal mines in Zuid-Limburg stimulated the establishment of the Dutch State Mines (DSM)

organisation. Since 1930, the DSM has also been active in the chemical sector. After World

War II, all products which had previously been imported were manufactured at home,

especially in the chemical industry. Exploratory borings for oil and gas, suspended during

World War II, were revived (Lubbers and Lemckert, 1980).

On the basis of the discovery of natural gas reserves in the northern province of

Groningen around 1960, it was estimated that 50 billion meter cubic (m3) of natural gas was

present. In 1965, the official estimates of natural gas reserves were around 1,500 billion m3

and two years later they stood at 2,200 billion m3 (Lubbers and Lemckert, 1980). After 1967,

the proven reserves of natural gas continued to increase. Throughout the 1960s, the Dutch

State realized the importance of natural gas extraction for the development of the economy

both at national and international levels (Lubbers and Lemckert, 1980: 87-117). The

Netherlands has developed a technologically high-quality oil and gas industry.

Central Bureau of Statistics (CBS, 2010) of the Netherlands, records suggest that the

share of natural gas exports was 17% of total exports of Netherlands during the time period

1960-69, but it had increased to 87% in 1970-79. These figures indicate increasing risk of

‘Dutch Disease’ during this time period. But the shares of natural gas exports in total exports

fell to 34%, 22% and 15% during the time periods1980-1989, 1990-99 and 2000-06,

respectively.

The natural gas rents (defined here as State income from gas revenues as a percentage of

GDP) are another measure of natural resource economic influence. The statistical data on

natural gas rents can be obtained from the World Development Indicators (WDI) (see CBS,

2010 and Table 1). Average values were calculated for five periods: 1960-69, 1970-79, 1980-

89, 1990-1999, and 2000-2006. The share of natural gas rents (as a percentage of GDP) were

higher in the time periods 1960-69 and 1970-79, respectively, as compared with the share of

natural gas rents in later periods.

In early 1986, a fall in oil prices led to a sharp decline in gas rents and most importantly,

a fundamental change in the medium term outlook for the European energy market (Radezki,

1999). For example, the import prices of gas to Western Europe declined from an average of

$3.7/mm BTU in 1984-1986 to $2.3 in 1987-1989, or by almost 40%, and remained at a

lower level for most of the 1990s (Radezki, 1999, p. 19). In addition, liberalization of East

Europe gas market (around 1990) and new opportunities for established gas market actors, as

well as, for new entrants provided a framework for the implementation of structural change

(Radezki, 1999) . Dependence on declining natural gas revenues for a country like the

Netherlands was becoming an unattractive prospect.

Natural gas indigenous production 1960-2006 by the Netherlands are summarised in

Table 2. The statistical data on these variables were collected from the Netherlands’ CBS in

millions of cubic meters. There initially was a rapid increase in total indigenous production of

natural gas, from 5125 (1960-69) to 76033 (1970-79). After the 1980s, the data indicates

there was little change in the rate of extraction of natural gas.

This time series suggests that the Netherlands’ experience could be divided into two

periods with a turning point in the mid-1980s. This paper explores the impacts of changing

policies on rate of extraction and associated contributions of rents from gas extraction plus

institutionalisation of more productive use of those rents, i.e. were these changes significant

in terms of the relative performance of the Dutch economy compared with the average

performance of OECD economies in the same period in terms of the economic indicators

most closely associated with Dutch Disease?

4. Comparing the Netherlands’ experiences with wider OECD experiences

Around 1960, the Netherlands was an economy with many of the structural

characteristics that might be expected of an OECD economy in the middle of the ‘long boom’

from 1950 to 1970. In terms of sectoral characteristics, a shrinking agricultural sector in

terms of percentage of GDP, a high technology manufacturing sector supplying both

domestic and global markets, and a growing service sector with increasing labour

productivity. In terms of human capital, increasing numbers of students were in tertiary

education. Macro-economic variables were healthy with a balance between international

inflows and outflows, stable exchange and interest rates, and low price inflation and

unemployment rates.

For the purposes of this paper, the OECD statistics cited are population weighted average

values from a variety of sources. These averages include the Netherlands economy’s own

values, but the Netherlands’ economy is regarded as making such a small contribution to

those average values that it can be treated as having an insignificant impact. The time series

are concluded in 2006 to avoid including the effects of the recent global economic crisis.

Earlier critical periods in the global economy are included as their effects are seen as being

statistically smoothed in subsequent years.

The Netherlands had a higher share of natural resource rents as compared with OECD

members during the period 1970-2006 (see Figure 1). This data was obtained from World

Development Indicators (World Bank, 2011), but this data is not available prior to 1970. The

lower share of natural resource rents in GDP after 1990 indicates the emergence of new, low

cost gas suppliers (especially Russia), but also a policy shift in the Netherlands (see Table 3

and Figure 2).

The average annual values for each sub-period were calculated, 1971-80, 1981-90, 1991-

00 and 2000-06. This statistical analysis suggests that the Netherlands has had higher levels

of production and exports of natural gas throughout the whole period, as compared with the

average for OECD Members (International Energy Agency, 2000: 2006). In consumption of

natural gas, the Netherlands has only slightly higher values as compared with OECD

Members. Regarding import trends, the Netherlands has a lower quantity of gas imports per

capita compared with the average of OECD Members from 1971-1980 and 1991-2000, but

this changes in 2000-06.

Overall, these statistics do suggest the Netherlands is an OECD outlier in terms of

national domestic natural resource availability. The question to be now addressed is whether

this outlier status has affected overall economic performance. More specifically, has this

status tended to increase the rentier aspect of the economy or is there evidence of the windfall

being used productively? This question is addressed by exploring five time series:

GDP per capita annual growth rates as an indicator of economic dynamism

Gross Capital Formation as an indicator of changing productivity

Saving rate as an indicator of sustainability

Tertiary education performance as an indicator of human capital development

The impact on ‘welfare’ expenditure as a key rentier response

The time series are divided into four sub-periods to facilitate visual comparisons between

longer run, trend values of the variables.

The first hypothesis to be tested is that a Dutch Disease resource curse would undermine

the relative dynamism of the Netherlands economy. Using GDP per capita annual growth

rates as an indicator of dynamism, The Netherlands economy grew slower than the average

growth rate of the OECD from 1970 to 1989, but grew faster in the 1990 to 1999 decade. But

the Netherlands’ relative growth rate fell behind the average again in the period 2000 to 2006.

While this comparison has some ambiguity in terms of testing the Dutch Disease hypothesis,

it is consistent with a Dutch Disease resource curse effect between 1970 and the late 1980s.

The GDP per capita time series does suggest a shift taking place in the 1980s that was

completed in the 1990s. One possible interpretation of this change in the comparative

dynamics of the Netherlands’ economy is that the initial response to the availability of gas

revenues was to let market forces determine the trajectory of the economy. This interpretation

receives some confirmation from the dramatic up-valuation of the Dutch guilder exchange

rate against the US dollar in 1970s then as soon as the post-WW2 fixed exchange rate regime

was abandoned, though the global crisis around 1980 did result in a substantial devaluation of

the guilder that lasted through the 1980s. This was followed by improvements in the external

current account and greater stability in the exchange rate in the 1990s (until the guilder

exchange rate time series ends with the adoption of the Euro around 2000 (see Figures 3 and

4).

The second hypothesis looks toward the use of windfall rents from gas extraction for

productive investment, rather than rentier consumption. The indicator used for this hypothesis

is Gross Capital Formation as a percentage of GDP. Comparing the Netherlands’

performance with the OECD average suggests very little difference across the whole period

from 1970 to 2006. This pattern is consistent with rejecting the hypothesis that the windfall

rents have been used productively to boost investment, but is consistent with an alternative

hypothesis that natural gas revenues have NOT actually reduced productive investment

compared with the OECD average performance.

Gross capital formation in the Netherlands as an indicator of possible productive use of

gas revenues appears unaffected compared to the wider OECD average by those revenues

throughout the whole period covered by this paper. This is despite the substantial difference

in savings corrected for depletion of natural capital, in which the Netherlands has a

substantially higher rate of net saving. It may be that the ‘surplus’ saving was being utilised

to purchase assets (including domestic properties) outside the Netherlands. In terms of the

rentier versus productive use of natural resource windfalls, the balance of the argument seems

to lie on the rentier side in terms of the absence of a clear comparative increase in national

productive capacity inside the Netherlands.

The third hypothesis is concerned with whether natural resource rents are used

sustainably. The indicator for this hypothesis is the comparative rate of saving, more

specifically the World Development Indicator series, available since 1990, of Genuine

Savings as a percentage of GNI that correct for depletion of natural capital. The Netherlands’

Genuine Savings rates were substantially and consistently above the OECD average rates for

the whole period 1990 to 2006. This pattern suggests an element of long duration

sustainability in the use of windfall rents post-1990 as savings which is consistent with a

macroeconomic income spreading ‘permanent income’ hypothesis in the use of windfall

income.

The fourth hypothesis focuses on human capital development and whether the natural

resource rents have been used to upgrade the skills of young people in the Netherlands. The

first indicator is the total public spending on education as a percentage of GDP which shows

an initial higher proportion for the Netherlands than the OECD average. But the proportion

and the difference decreases from 1970 to 1999. Though the Netherlands’ proportion rises in

2000 to 2006, it is matched by an overall rise in the OECD average. This performance

suggests the windfall revenues have not significantly benefitted human capital development.

A second insight into human capital development comes from the pattern of subjects

chosen by Netherlands’ students (see Figures 5 and 6). Dividing education between

production and consumption choices is controversial, but the data shows an initial surge in

numbers of people studying the ‘more consumptionist’ humanities followed by a leap in

social sciences, More ‘socially and environmentally responsible’ numbers of people studying

science and engineering do rise between 1970 and 1989 but then stagnate from 1990 in

absolute terms, though vocational education shows a steady increase over the whole period.

Overall, it does appear that development of more technologically productive, high level

human capital through the education system may not have benefitted from the allocation of

windfall natural resource revenues seen through the decisions made by young people in

choosing subjects to study.

The interpretation of the supply side of education is even more complex. Public spending

on education in the Netherlands was substantially higher that the OECD average in the 1970s

when students were choosing to study humanities, but fell in comparative terms in the 1980s

and 1990s with some revival after 2000. If education is regarded as an activity connected to

labour productivity in work, then the gas revenues seem to be having a diminishing effect

over time. Student choices of subjects and the relatively weak performance in the sciences are

consistent with less productive use of gas revenues, as far as skilled labour in the agriculture

and manufacturing sectors is concerned. But the students’ choices are consistent with

increases in productivity in the services sector with its complex mix of highly tradable and

less tradable commodities. As in Gylfason (2001a), it is difficult to draw any firm

conclusions about the impact of gas revenues on human capital development from the

indicators used in this paper.

The fifth hypothesis is that Dutch Disease resource curse will be reflected in greater use

of resources for providing incomes to people whose labour force statuses are either

economically inactive or economically active but unemployed. Arguably this is the clearest

manifestation of a Dutch Disease resource curse rentier effect in a higher income economy,

even if the increased ‘leisure’ is involuntary for some of the people not working. The

indicator used to test this hypothesis is the ‘social contribution’ as a percentage of GDP as

reported in the World Development Indicator’s (2011) and Bos (2008) statistical series. The

Netherlands’ social contribution was a much higher percentage of GDP than the OECD

average in the period 1970 to 1990. The higher percentage is maintained in the period 1990 to

2006, but against a background in which the OECD average was rising and the Netherlands’

falling significantly. As with the GDP per capita statistics, this pattern suggests a contrast

between the first and second halves of the period.

More directly in terms of a tendency to rentier use of gas revenues is the expenditure on

‘social contributions’. Social contributions bring together all the benefits received by people

that might be considered as providing disincentives to work at all or work longer hours. They

therefore reflect the aggregate ‘leisure preferences’ in society and how far people can ‘afford’

to trade-off between being rentiers and being workers. The Social Contribution time series for

the Netherlands suggests a stronger rentier element than in the wider OECD. But there is

convergence from around 1990. Arguably this is the strongest indicator of both the presence

of a stronger rentier affect in the earlier period and the emergence of a greater production

affect from around 1990.

5. From ‘curse’ to ‘miracle’ and distribution to production: reflections on the political

economy of Netherlands’ ‘extractivist’ experiences

Overall, statistical indicators confirm the roles of natural resources and associated rents

are higher in the Netherlands compared with the average for OECD countries. This indicates

a potential for resource curse or blessing. The Netherlands had higher levels of production,

consumption and exports of natural gas as compared with the average for OECD countries

throughout the period explored in this paper. The analysis in this paper rests on the claim that

this higher natural resource availability is an important determinant of national economic

performance and the time series in the selected macro-economic indicators. The interpretive

reflections presented in this paper are in the nature of exploratory data analysis and only more

robust qualitative conclusions will be presented here.

There appears to have been a change in the role of gas revenues around 1990, already

indicated in the comparison of natural resource rents shown in Figure 1. Figure 2 shows a

halving of revenues from between four to five percent of GDP to less than two percent

through the 1990s into the early years of the 2000s.

Since the mid-1990s, a third of public sector natural gas revenues have gone into a

special fund known as ‘Economic Structure Enhancing Fund’. The purpose of the fund is to

improve the economic infrastructure of the Netherlands, such as high-speed rail links and

major highways. The fund expenditure is based on money received from natural gas revenues

(Blöndal and Kristensen, 2002 and Bos, 2008). The objective of the fund is two-fold. The

first objective is to remove part of gas revenues from political public sector budget debates.

The second objective is to guarantee that significant amounts of money are spent on

improving the country’s economic infrastructure.

The findings in this paper indicate a shift from an extractivist to more socially and

environmentally responsible focus starting in the mid-1980s. This is supported by Stijns

(2005) argument that the ability to turn a resource curse into a blessing depends on the nature

of the learning process. Building on a pre-existing contextual national history of economic

and political development, Dutch governments appeared to have shown a capacity to learn

about sustainability (through reduction in the rate of extraction) and reducing the comparative

rate of growth of welfare expenditure and increasing genuine savings. The push towards a

transition from ‘welfare to work’ may be creating a ‘Dutch Miracle’ (in the language of

Becker, 2000, p.230) rather than ‘Dutch Disease’.

The Netherlands have greater advantage over many other natural resource abundant

economies due to its context of a strong pre-existing manufacturing sector, a long duration

historic institutional process giving political stability, low level of corruption, and relatively

good governance (according to International Standards, see Table 4), plus relative autonomy

from the more aggressive aspects of the global system (arguably protected by being a

member of the European Economic Community, now the European Union). Nevertheless, the

Netherlands’ government took a period of about twenty years to change direction, albeit more

like an oil tanker than a racing yacht, towards a production focus and away from a

extractivist/rentier tendency.

In terms of understanding the agency that produced this change of direction, it is

interesting to investigate whether the turning point in the 1980s had a dramatic political

manifestation. There were four national Parliamentary elections in the Netherlands in the

1980s in 1981, 1982, 1986, and 1989 (Horstman and van Deth, 1997). The turnouts

fluctuated between 87.0% (1981) and 80.3% (1989) with no clear downward or upward trend

(voting was made voluntary rather than compulsory in the 1960s). Valid votes were all over

99% of the total votes cast. These statistics suggest a substantial level of political

commitment to the Parliamentary system around the time of the change in strategy, but no

increase in overall political engagement among Netherlands’ citizens.

The distribution of votes between the political parties was dominated by two parties, the

left leaning PVDA and the centre right CDA. The PVDA share of the vote varied between

28.3% (1981) and 33.3% (1986) while the CDA share varied between 29.4% (1982) and

35.3% (1989). The two parties together received an increasing share of the vote during the

1980s rising from 59.1% in 1981 to 67.2% in 1989. But to form a bare majority government,

both parties required at least one other party to join a coalition. While it was easier to form a

centre right coalition in the 1980s, there is no comparison with the neighbouring country

across the North Sea where ideologically committed neo-liberals led by Margaret Thatcher

dominated politics throughout the 1980s (and arguably ‘wasted’ a lot of the UK oil revenues

on unemployment ‘benefits’ giving a whole generation experience of unwanted ‘leisure’ –

only the accident of chronology prevented resource curse being labelled ‘UK’ rather than

‘Dutch’ disease!).

So the change in direction was not as a result of a dramatic shift in politics. A plausible

explanation in the Netherlands’ context is that the change was driven by an institutionalist

technocratic tendency in Dutch national decision-making. To personalise this argument, it is

useful to compare the thoughts of post-WW2 Dutch economist Jan Tinbergen with his

contemporary USA economist, Milton Friedman. Tinbergen developed elaborate

institutionalist planning models while Friedman was a fierce advocate for neo-classical ‘free’

market forces. It is difficult for these authors to detect the discourse (in both argumentation

and evidential aspects) that produced the change, partially due to language and partially due

to the discreet nature of Dutch ‘public’ policy processes (but see Bos, 2008; Wierts and

Schotten 2008 for published accounts of policy processes in the Netherlands).

Overall, a benign institutional economics approach, including a learning process, seems

to be most appropriate for understanding the impact in the Netherlands of its natural resource

windfall. The Netherlands’ experience suggests that without positive pro-socially and

environmentally responsible policy actions, market forces did tend to produce a more rentier

economy, but this was off-set, if not eliminated, by policy actions, albeit delayed and not

inevitable. Comparing Netherlands and UK experiences indicates the importance of

understanding long duration institutional history in seeing why some countries experience

can be seen through an optimistic institutional economics lens and others need to be seen

through a more pessimistic political economy lens.

6. Possible lessons for green political economy from the history of natural resource

‘extractivism’ in the Netherlands.

The empirical evidence derived from economic indicators in this paper suggests, while

the Dutch contracted a ‘disease’ and suffered from it for a significant period, ‘Dutch Disease’

did not prove fatal and, eventually, there was a substantial, if uneven, recovery. But there was

little or nothing in this recovery due to the operation of liberalising, ‘free’ market forces. For

almost twenty years, the pressure of market forces was in the direction of creating a ‘rentier’

based economy in which aggressive ‘extractivist mining’ of fossil gas and associated

greenhouse gas emissions would create greater ‘leisure’ and offer less employment to the

people of the Netherlands.

This tendency to greater ‘leisure’ is distributionally conditioned by the political economy

historical context of the national economy in its specific global context. In the Netherlands,

there was a pre-existing ‘welfare’ state that provided institutional channels for resources to

reach a large proportion of the population. This included the formal unemployed displaced by

now being uncompetitive in terms of foreign exchange measured productivity in the global

economy, even though their physical productivity was unchanged. In the Netherlands, these

people were now ‘freed to enjoy’ greater leisure on welfare benefits. Young people were also

‘freed’ to choose patterns of education that were not necessarily conducive to engaging in

manufacturing activity. And higher income people are ‘freed’ by access to cheap foreign

exchange to place their savings outside the national economy.

The sector most clearly in favour of ‘free’ market forces operating to aggressive natural

resource extraction is the fraction of capital that actually does the extraction, assuming its

unit cost of extraction is less than the global average unit cost. This fraction of capital can

exercise political influence through contributing from its ‘excess’ profits/economic rents to

public sector revenues and wider ‘lubrication’ of the political/bureaucratic system.

Governments will also be sensitive to the pricing of commodities derived from natural

resources that have linkages to every sector of the economy. Though the sector may employ

relatively few national employees, inter- and intra-corporate global linkages can help put

international pressure on national governments wishing to limit rates of natural resource

extraction.

The main indirectly effected sector which may have ambivalence about the costs of

unbridled operation of ‘free’ market forces in natural resource extraction is the financial

services sector and its offshoots in property financing. A comparison with the UK is useful in

which the enormously powerful City of London banking sector with its client national

politicians presided over the collapse of UK manufacturing and coal-mining industries as UK

fossil oil revenues exploded around 1980 and subsequently the public sector paid enormous

amounts in unemployment ‘benefits’ throughout the 1980s.

The possible lessons for green political economy from the Netherlands’ experience point

towards understanding the political economy of how patterns of fractions of capital interact in

creating the elite discourse on the role of natural resource extraction. From a political

economy perspective, there are fractions of capital that support or oppose the operation of

free market forces in deciding the role of natural resource extraction in the national economy.

But more empirically, these positions are best seen as based on opportunistic material

interests rather than collective fundamental beliefs in free markets or deep nationalist

sentiments.

Natural resource extraction redistributes opportunities to make profits. Significant

financial and knowledge exit and entry costs can obstruct capital moving towards the sectors

offering greater profitability and thus convert the abstract concept of fractions of capital into

sustained political lobbies. But a fraction of capital lobbying for the operation of free market

forces in natural resource extraction does not guarantee the same fraction will not lobby for

state intervention on a subject where it feels free market forces operate against its material

interest, much to the confusion of politicians and intellectuals of a more ideological

conviction.

This naturally brings the argument to the role of formal political processes in the

Netherlands’ experience. The Netherlands’ governance regime historically emerged in the

process of the hegemonic ‘western’ development of ‘democracy’ as national ‘universal’ adult

franchise, periodic elections with individualised secret ballots, and multiple political parties

open to mass membership. Arguably the Netherlands can be seen as an advanced form of

such a governance regime in formal terms. But, like all such regimes at the national level,

there is a long duration historical process that colours it in practice. Netherlands’ civil society

emerged from its period of Spanish colonisation with deep religious divisions between

Roman Catholics and Protestants. The long term political ‘solution’ was to accommodate this

division in an informal acceptance of ‘pillars’ in which societal divisions were given

considerable organisational autonomy with national unity being constructed in the pediment

on top of the pillars. This construction, though weakening in the last half century, has

produced continuity and stability in government able to tolerate frequent circulation of

political parties and their usually temporary leaders. This tolerance of political circulation can

be seen as logically associated with significant technocratic influence on governance

deliberations.

Therefore it is not surprising the advent of a large natural resource revenue windfall was

initially treated as a populist opportunity to increase public expenditure, and thus ensure the

popularity of whichever political parties happened to be in government. Increasing aggregate

effective demand could temporarily placate all fractions of capital. In terms of mass political

mobilisation, the lobby on behalf of the unemployed was also muted by increasing welfare

benefits and cheaper imported commodities. Such rational, if short term, populist public

policy behaviour persisted and was not challenged in national elections.

Alongside possible increased lobbying from those fractions of capital experiencing

pressure on profitability, especially larger scale manufacturing, Netherlands’ politicians were

probably being faced with projections of unsustainable increases in the public sector welfare

bill for the public sector by their official and independent ‘advisers’. More widely, this raises

the issue of the role of evidence based, technocratic influences on public policy institutional

deliberative processes.

It is also likely that more persistent unemployment was being associated with public

order challenges, not by mass public demonstrations, but more by day to day street level

‘anti-social untidiness’ offending ‘ordinary citizens’ in major cities. Though election results

did not suggest any demands for radical change, Dutch society’s pillars can informally

channel discontent from local to national levels. Using an appropriate metaphor, responding

to the sound of surf ahead in a dense fog, the Netherland’s governance ‘Dutch public policy

oil tanker’ slowly turned from a distributional focus towards a more production oriented

focus.

For Latin America, the lesson may be that the more popularly accountable the

governance regime, then the more likely that a populist distributive reaction will persist. Thus

the pressures of open market forces and short-term political expediency will reinforce each

other. But popular accountability is not a necessary or sufficient condition for a persistent

extractivist focus. A ‘grabber friendly’ regime (Mehlum et al, 2006) is also likely to show

extractivist persistence as long as a mixture of corruption and coercion permit resources to be

‘grabbed’.

The key difference is whether a more accountable governance regime will be more likely

to make a ‘socially and environmentally responsible’ turn earlier, and more determinedly,

than a less accountable governance regime? More accountable regimes may be more able to

accept lobbying from ‘dissident’ fractions of capital faced with falling profitability as a

consequence of natural resource extraction. Such regimes may also be more sensitive, and

less coercively reactive, to ‘middle class’ concerns that civil society is becoming more

disorderly as increasing numbers of young ‘descamisados’ acquire English Premier League

club football shirts and find other more anti-social ways of signifying their discontent with

their adverse incorporation in ‘mainstream’ society.

The Netherlands’ experience does not provide a how-to-do, self-help, universally

applicable manual on how to survive catching ‘Dutch Disease’. The Netherlands’ experience

demonstrates how open market forces tend to drift the economy in a rentier direction and how

rational political decisions to do nothing initially support this drift by focusing on

distributional processes. Political economy analysis of the Netherlands’ experience illustrates

the complex pattern of forces that may, and only may, operate later to change direction

towards a more socially and environmentally responsible focus. This analysis may be useful

as a framework for understanding aggressive extractivism, though it offers no optimistic

inevitability in turning a ‘resource curse’ into a ‘blessing’ with reduced, more sustainable

rates of extraction.

REFERENCES

Al-Kasim, F., T. Soreide, and A. Williams (2013) ‘Corruption and reduced oil production: an

additional resource curse factor?’, Energy Policy 54: 137-147.

Auty, R. (2001a) 'The Political Economy of Resource-Driven Growth', European Economic

Review 45(4): 839-846.

Auty, R. M. (2001b). (Ed.) “Resource abundance and economic development”, Oxford:

Oxford University Press.

Auty, R.M. (2007a) 'Natural Resources, Capital Accumulation and the Resource Curse',

Ecological Economics 61(4): 627-634.

Auty, R.M. (2007b) 'Rent Cycling Theory, the Resource Curse, and Development Policy',

Developing Alternatives 11(1): 7-13.

Becker, U. (2000) 'Welfare State Development and Employment in the Netherlands in

Comparative Perspective', Journal of European Social Policy 10(3): 219-239.

Blöndal, J.R. and J.K. Kristensen (2002) 'Budgeting in the Netherlands', OECD Journal on

Budgeting 1(3): 43-80.

Bos, F. (2008) 'The Dutch Fiscal Framework: History, Current Practice and the Role of the

Central Planning Bureau', OECD Journal on Budgeting 8(1): 7-48.

???? Boschini, A., J. Pettersson, and J. Roine (2012) ‘The resource curse and its potential

reversal’, World Development 43: 19-41.

Bravo-Ortega, and C., J. De Gregorio (2005) The Relative Richness of the Poor? Natural

Resources, Human Capital, and Economic Growth. Vol. 3484. World Bank, Latin

Ameria and the Caribbean Region, Office of the Chief Economist.

Brown, J. C. (1985) ‘Why foreign oil companies shifted their production from Mexico to

Venezuela during the 1920s’, The American Historical Review, 90(2): 362-385.

Brunnschweiler, C. N. (2008) ‘Cursing the blessings? Natural resource abundance,

institutions, and economic growth’, World Development 36(3): 399-419.

Cameron, J. and T. P. Ndhlovu (2013) 'Economics and 'Excess': Implications for

Understanding and Combating Climate Change', International Journal of Sustainable

Economy 5(1): 15-35.

Central Bureau of Statistics. (2010). Data retrieved May 2, 2012, from StatLine database.

URL http://statline.cbs.nl/ (accessed May 2012).

Cole, K., J. Cameron, and C.B. Edwards (1992) Why Economists Disagree, Second Edition, London, Longman.

Corden, W.M. (1984) 'Booming Sector and Dutch Disease Economics: Survey and

Consolidation', Oxford Economic Papers 36(3): 359-380.

Davis, G.A. and J.E. Tilton (2005) 'The resource curse', Natural Resources Forum, Wiley

Online Library pp 233-242.

Di John, J. (2007) 'Oil Abundance and Violent Political Conflict: A Critical Assessment',

Journal of Development Studies 43(6): 961-986.

Gylfason, T. (2001a) 'Natural Resources, Education, and Economic Development', European

Economic Review 45(4): 847-859

Gylfason, T. (2001b) 'Lessons from the Dutch Disease: Causes, Treatment, and Cures',

Institute of Economic Studies-University of Iceland.

Gylfason, T. and G. Zoega (2002) 'Inequality and Economic Growth: Do Natural Resources

Matter?' CESifo Working Paper Series No. 712

Horstman, R. And J. W. van Deth (1997) Dutch Parliamentary Election Studies Data Source

Book 1971-1989, ICPSR Version 6744, Inter-University Consortium for political nand

Social Research, Michigan, Ann Arbor

Hutchison, M.M. (1994) 'Manufacturing Sector Resiliency to Energy Booms: Empirical

Evidence from Norway, the Netherlands, and the United Kingdom', Oxford Economic

Papers: 311-329.

International Energy Agency (2000) Natural Gas Information 2000. Paris: Organisation for

Economic Co-operation and Development.

International Energy Agency (2006) Natural Gas Information 2006. Paris: Organisation for

Economic Co-operation and Development

International Monetary Fund (2012)l Balance of Payment Statistics, BOPS. Data retrieved

June 5, 2012, from International Monetary Fund (IMF) Online database (accessed

June 2012).

Jerome, A., S. Adjibolosoo and D. Busari (2005) 'Addressing Oil Related Corruption in

Africa. Is the Push for Transparency enough?’ Review of Human Factor Studies 11(1):

7-32.

Kaufmann, D., A. Kraay and M. Mastruzzi (2010) 'The Worldwide Governance Indicators:

Methodology and Analytical Issues', World Bank Policy Research Working

Paper (5430).

???? Lubbers, R.F.M and C. Lemckert (1980) 'The Influence of Natural Gas on the Dutch

Economy', in Griffiths, R.T. (eds). The Economy and Politics of the Netherlands since

1945 Century. Nijhoff.

???? Mahdavy, H. (1979) ‘The patterns and problems of economic development in rentier

states: the case of Iran’ in Cook, M. A. (ed), Oxford University Press

McKinley, Terry (2008). ‘Resource Abundance: A Curse or a Blessing?’. In Terry McKinley

and Katerina Kyrili (eds). The Resource Curse. Centre for Development Policy Research

(CDPR) Development Digest No. 1, December, SOAS, London.

Mehlum, H., K. Moene, and R. Torvik (2006) ‘Institutions and the resource curse’, The

Economic Journal 116: 1-20.

Murshed, S.M. (2008) 'Development Despite Modest Growth in the Middle East’, Review of

Middle East Economics and Finance 4(3): 1-31.

Murshed, S.M. (2011) 'What Turns a Blessing into a Curse? the Political Economy of Natural

Resource Wealth', The Pakistan Development Review 46(4): 351-377.

Murshed, S. M. and L. A. Serino (2011) ‘The pattern of specialization and economic growth:

the resource curse hypothesis revisited’, Structural Change and Economic Dynamics 22:

151-161.

???? North, D. And R. Thomas (1973) The Rise of the Western World: A New Economic

History, Cambridge University

Raghuram, G. R and A. Subramanian (2011) ‘Aid, Dutch Disease, and manufacturing

growth’, Journal of Development Economics, 94: 106-118

Radetzki, M. (1999) 'European Natural Gas: Market Forces Will Bring about Competition in

any Case', Energy Policy 27(1): 17-24.

Sachs, J.D. and A.M. Warner (1995a), Natural resource abundance and economic growth,

NBER Working Paper No. 5398.

Sachs, J.D. and A.M. Warner (1995b), Economic convergence and economic policies. NBER

Working Paper No. 5039 (Also Reprint No. r2002)

Sachs, J.D. and A.M. Warner (1999) 'The Big Push, Natural Resource Booms and Growth',

Journal of Development Economics 59(1): 43-76.

Sachs, J.D. and A.M. Warner (2001) 'The Curse of Natural Resources', European Economic

Review 45(4): 827-838.

Sampson, A. (1975) The Seven Sisters, London, Coronet Books, Hodder and Stoughton

Stijns, J.P.C. (2005) 'Natural Resource Abundance and Economic Growth Revisited',

Resources Policy 30(2): 107-130.

Van der Ploeg, F. (2011) 'Natural Resources: Curse Or Blessing?’ Journal of Economic

Literature 49(2): 366-420.

Van der Ploeg, F. and S. Poelhekke (2009) 'Volatility and the Natural Resource Curse', Ox-

ford economic papers 61(4): 727-760.

Wierts, P. and G. Schotten (2008), Dutch natural gas revenues and fiscal policy: Theory ver-

sus practice. Netherlands Central Bank, Research Department, DNB Occasional Studies 605

World Bank. (2011). Data retrieved April 20, 2012, from World Development Indicators

Online (WDI) database.