This appendix first provides some further comments on the operational definition of retrenchment. It then explains how the budgetary figures have been calculated. Finally, it provides two examples of such calculations.

First, not all of the changes implemented in the social-security schemes studied are retrenchments. There are several examples of improvements of the schemes and these have been included and studied in the same way as retrenchments. Not including improvements would have lead to very misleading conclusions about differences in extent of retrenchment.

Second, a retrenchment can be made up of more than one single change in a social-security scheme. A number of changes may be tied together, all of which are not necessarily retrenchments. Therefore, the overall impact of changes carried through as one has been used as the baseline. This implies that a retrenchment, which is part of an overall expansion of a social-security scheme has not been analysed as a retrenchment. Consequently, changes in a social-security scheme altering the distribution of benefits between groups and of which the overall budgetary impact is zero have also not been considered retrenchment, notwithstanding the fact that such changes may have imposed considerable losses on some groups of claimants. Such changes are more appropriately viewed as a restructuring reform with greater targeting.

Third, budgetary implications in relation to changes in one social-security scheme may affect another social-security scheme, and this must be taken into account. Significant cutbacks in one social-security scheme are not likely to be of much interest to the electorate if they are simply compensated by improvements in another scheme. Therefore, when calculating the figures, the effects on other schemes have been taken into account.

A related problem is to delimit social-security schemes from the tax system. Almost all of the social-security benefits examined in this study are taxable income, so the effects of retrenchment will partly be offset by the tax system. Yet, such effects have generally been disregarded for three reasons. First, due to the complexity of the tax systems in the two countries, a calculation of net-effects is almost impossible. Second, the disturbing effects that this may have for the conclusions concerning retrenchment seem limited, as almost all the benefits are taxable income.

Following Pierson's seminal work (1994; 1996), it has become conventional wisdom within the literature on welfare-state retrenchment that retrenchment politics are fundamentally different from expansion politics. This indicates that the scholarly debate about welfare-state retrenchment would be markedly different from the debate about welfare-state expansion, which in several respects is also the case (Green-Pedersen & Haverland 2002). Nevertheless, as will become clear below, the retrenchment debate has not started from scratch. It has been influenced by the preceding and ongoing debate about the growth of the welfare state. Therefore, it is necessary to start with a short introduction to the voluminous literature on welfare-state expansion.

Theories of welfare-state expansion

In the following, a brief sketch of the main positions with regard to the growth of the welfare state will be given.1 This should provide the reader with the necessary background for understanding the retrenchment debate. In general terms, theoretical arguments about the expansion of the welfare state fall into three groups; namely structural or functional theories, theories emphasising the role of politics, especially party politics, and state-centered or institutional theories.

Structural theories all link the emergence of the welfare state to structural changes in society such as modernisation, democratisation, economic growth, or changes in labour market structures (Wilensky 1975; Flora & Alber 1984; Iversen & Cusack 2000, cf. also Pierson 1991: chap. 1 & 2). These societal changes have lead to the emergence of the welfare state. The main aim of these theories has been to explain why all Western countries developed some kind of welfare state. Less attention has been paid to comparative differences. Still, some ‘structural studies’ have focussed on variation among welfare states. By way of example, Cameron (1978) and Katzenstein (1985) emphasize the degree of openness of the economy as a factor leading to different welfare-state developments. Nevertheless, the main impact of the ‘structural theories’ has been to highlight the existence of certain prerequisites for welfare-state development such as democratisation, modernisation etc.

The second group of theories all emphasize the role of politics, especially party politics, in the expansion of the welfare state. With the structural theories, these theories share the ambition of explaining the growth of the welfare state in general.

The development of modern welfare states is an issue that has attracted enormous attention within social science. One could easily fill an impressive library with books and articles about the issue. The reasons for this interest are straightforward. The welfare state is a highly important phenomenon in Western societies. It strongly influences the life of nearly the entire population in the Western world. Thus, researchers interested in different subjects such as gender relations (e.g. Lewis 1992; Orloff 1993) or poverty (e.g. Korpi & Palme 1998) automatically come to focus on the welfare state.

Until about 20 years ago, studying the development of modern welfare states was synonymous with studying their growth. This no longer applies. Welfare-state retrenchment has been on the political agenda of most of the rich oecd countries for the last 20 years, and almost all countries have experienced welfare-state retrenchments (Stephens, Huber, & Ray 1999). This has been part of a larger attempt of Western societies to adjust their models of welfare capitalism to a changed economic environment (cf. Kitschelt et al. 1999; Regini 2000; Scharpf & Schmidt 2000a). Evidently, the recent development of modern welfare states has also affected the scholarly interest. Research into welfare-state growth is still ongoing (e.g. Hicks 2000; Iversen & Cusack 2000; Mares 2000; Huber & Stephens 2001a). However, starting with Pierson's (1994; 1996) groundbreaking studies of welfare-state retrenchment in primarily the United States and Britain, a scholarly debate about retrenchment has gained momentum.

This book is an investigation of retrenchment in Denmark and in the Netherlands during the period 1982 to 1998, focussing on retrenchment of old-age pensions, unemployment benefits, and disability pensions/early retirement benefits. The general aim of the book is to contribute to the scholarly debate about welfare-state retrenchment in a number of ways. The main aim is to add to the knowledge of the factors causing variation in retrenchment between countries and across social-security schemes. Here, the argument of the book is that politics is the main factor causing variation. In this way, the book differs from most other writings on retrenchment that follow the new institutional wave within political science in general, highlighting the role of institutional factors.

This chapter aims at two things. First, at briefly summarising the results of both parts of the empirical investigation. This will be done by discussing each of the six hypotheses developed in chapter 4. Second, a number of possible quibbles and alternative explanations will be discussed. Chapter 2 presented a number of theoretical propositions about variation in welfare-state retrenchment. The general aim of the research design was to be able to rule out such alternative explanations. However, this was as always only possible to some extent, leaving a number of plausible alternative explanations that deserve discussion.

The empirical support for the framework on balance

The first hypothesis was that there would be no sweeping retrenchment. Chapter 5 revealed support for this hypothesis with the qualification that defining ‘sweeping’ is difficult indeed. The underlying argument was the idea of the politics of retrenchment as one of blame avoidance. The Danish as well as the Dutch case supported this idea. In both countries, retrenchment proposals were met with demonstrations and protests. Several of the governments also experienced electoral punishment for retrenchment. However, both cases also demonstrated that governments are not necessarily punished for retrenchment.

The second of the six hypotheses suggested the implementation of more retrenchment in the Netherlands than in Denmark, based on the different systems of party competition. This hypothesis was confirmed in chapter 5 and chapter 7 showed that a consensus on retrenchment policy did in fact not emerge in Denmark until after the Social Democrats regained power in 1993. As long as they were in opposition, they had no incentive to give up their ownership of the welfare-state issue by engaging in welfare-state retrenchment. The Dutch case demonstrated how the pvda was ‘forced’ to abandon its ownership of the welfare-state issue through the 1980's to become an acceptable government partner for the cda. Thus, it does seem that the different systems of party competition can explain the variation between the two cases. However, it goes without saying that the framework cannot explain all the details in the party competition surrounding retrenchment. In the Dutch case, the framework would have, for instance, predicted the PVDA to choose a centrist course more rapidly than the party in fact did, and it is also surprising that the cda gave away part of its pivotal power when it declared the VVD its preferred government partner before the 1986 election.

This chapter aims to draw a Danish parallel to the Dutch case study of the previous chapter. The chapter begins with a short section on Danish politics and the welfare state in the years preceding 1982, especially the period 1973 to 1982. A second section explores the period of right-wing rule from 1982 to 1993. The third section addresses the period of social democratic-led governments from 1993 to 1998. Finally, some general conclusions are given about the Danish case.

Landslide election and fumbling socio-economic policy

Danish politics have historically been dominated by a left wing and a rightwing bloc. The former has largely been identical to the Social Democratic Party (Socialdemokratiet). The latter consists of the Conservative Party (Det Konservative Folkeparti) and the Liberals (Venstre), an agrarian party gradually transformed into a liberal party. The Social Liberals (Det Radikale Venstre) held the balance of power and hence decided which bloc should govern the country. For the first three post-war decades, primarily the Social Democrats were in power. At times, the Social Democrats ruled in majority governments including the Social Liberals, and at other times they formed single-party minority governments (Pedersen 1987; Damgaard 1992; Christensen 2000). The only major change to this pattern before 1973 was the establishment of the non-communist, moderate left-wing party, the Socialist People's Party (Socialistisk Folkeparti, SF), and its entrance into parliament in 1960. During the two periods 1966-1968 and 1971- 1973, the Social Democrats were able to rely on the support from sf. The Social Liberals then chose to support the right-wing bloc (op. cit.).

The 1973 election changed this situation as a number of new parties gained seats in parliament, among which the largest one was the right-wing Progress Party (Fremskridtspartiet) gaining around 16% of the votes. Among the other newcomers were two centre parties, the Centre Democrats (Centrumdemokaterne, CD) and the Christian People's Party (Kristeligt Folkeparti). In addition, the Communist Party (Danmarks Kommunistiske Parti) and the Justice Party (Retsforbundet) regained representation (Pedersen 1988). The 1973 earthquake election did not change the bloc nature of Danish party competition, yet it complicated Danish politics.

This book is an investigation of retrenchment in Denmark and in the Netherlands during the period 1982 to 1998, focussing on retrenchment of old-age pensions, unemployment benefits and disability pensions/early- retirement benefits. The general aim of the book is to contribute to the scholarly debate about welfare-state retrenchment in a number of ways. The main aim is to add to the knowledge of the factors causing variation in retrenchment between countries and across social-security schemes. Here, the argument of the book is that politics is the main factor causing variation. In this way, the book differs from most other writings on retrenchment that follow the new institutional wave within political science in general, highlighting the role of institutional factors. An additional aim of the book is to add to the empirical knowledge about retrenchment. As argued by Alber (1996: 3), ‘… systematic studies of retrenchment processes in a comparative perspective are still thoroughly missing.’ This book aims at filling a small part of that vacuum.

However, this book should not only interest scholars engaged in the debate about welfare-state retrenchment. Welfare-state retrenchment is part of the broader effort of Western governments to adjust their models of welfare capitalism to a new economic environment. Welfare-state retrenchments are probably the most controversial part of this effort, and this book therefore contributes to the understanding of different country's capacity to adjust their models of welfare capitalism. Further, Denmark and the Netherlands have come to be seen as two of the success stories with regard to this adjustment process. Scholars are thus debating both a Dutch and a Danish miracle (cf. Visser & Hemerijck 1997; Schwartz 2001; Green-Pedersen 2001). The book draws attention to the role that party politics has played in the adjustment processes within the two countries. This includes both the development of the party systems in the two countries and the situation of the Social Democratic Parties. The discussion of the latter factor also contributes to the wider debate about the dilemmas of European Social Democracy.

This book is published on the occasion of the celebration of the 250th anniversary of the Holland Society of Arts and Sciences. The founders of our Institution were – as could be expected in the middle of the 18th century – deeply convinced that science, and perhaps also to some extent the arts, should play a significant role in the prosperous development of the town of Haarlem and its surroundings in Holland. Several comparable ‘Learned Societies’ already existed in Europe and America, but our Society was the first to be established in the Netherlands.

Since the foundation of the Holland Society the enlightened optimism about the impact of scientific work has changed to a more prudent vision. Nevertheless, we undertook this project following the traditional aim of our Institution to advance science for the benefit of society, by selecting as its subject the interaction between humanity and the biosphere. This subject is of course not new. Through the last decades many different and richly faceted studies have been undertaken and articles and books that shed new light on this interaction are published almost daily. However, many questions remain unanswered and it seems that the number of issues and questions increases rather than diminishes.

It is obviously beyond the scope of one limited project – limited in time, manpower and budget – to give satisfactory answers to many of these questions. But we think that this book will give a new and deeper understanding by means of the approach taken and the inclusion of recent and innovative points of view. As the title of the book suggests, the authors have made extensive use of maps, both historical and contemporary, while exploring less known or unknown territory. These maps and many local and regional narratives do not only illustrate this study, they also facilitate our view of the complex interaction between humans and the environment.

In this book sustainable development is not dealt with as a rather static desirable track between the present and the future, but as part of a long-term, dynamic and evolutionary process of the co-existence of humanity and its environment that started many millennia ago.

Introduction

Given the limited time and the clearly overly large ambition to sketch Mappae Mundi for all places and times in human history and pre-history, we now take a big leap and focus on the last 400 years. Around the year 1000 AD about two out of every three human beings lived in Asia and some 70% of economic activity took place there (Maddison 2001). The larger part lived in the densely populated plains of China and India. By the year 2000 AD still about 60% of the world population lived in Asia but its share in economic activity had declined to 37% – at least when measured as the gross domestic product (GDP). Probably the dominant phenomenon of the intervening millennium has been the expansion of European peoples and their activities to most of the ‘empty’ regions of the world. This is the topic of Chapter 10. In this chapter we attempt to give an impression of how the two-thirds of the human population in Asia was doing, with an emphasis on population-environment interactions.

There are two reasons why the population-environment dynamics in Asia are important. Firstly, as the anthroposphere is expanding extensively and intensively, the population in Asia – and its socio-economical and cultural patterns – will have an enormous impact on how the world population will move to the next ecological regime because of its sheer momentum and its relatively low level of economic activity. Secondly, the population in several parts of Asia was confronted in this period with the limits of the carrying capacity of the environment. This manifested itself in famines and revolts whenever and wherever the socio-economic and institutional mechanisms used to sustain the rather large population densities were inadequate. As such they provide valuable insights in the various ways people coped – or failed to cope – with such pressures in these regions and probably in similar situations earlier and elsewhere. Our accounts will be about India, Indonesia and Siberia; we failed to receive one about China. We are fully aware that this eclectic approach with regard to this important part of our Mappae Mundi is far from satisfactory. However, even finding these tiny pieces of demographic and environmental history turned out to be a tedious task.

Introduction

Human-induced modification of the environment started with the use of fire,much earlier than any form of agriculture. Fire opened up the land for hunting and early forms of horticulture and agriculture and pastoralism. The subsequent process of agrarianization has already been described in Chapter 2. The agrarian regime saw a series of tools and practices, including the adaptation of new crops and the domestication of animals. Animal domestication and a more sedentary existence influenced population growth. Parts of the natural environment – good soil, sources of water and wood, mineral deposits – became a ‘resource’. New forms of social organization came into existence. Pockets of natural landscapes became dominated by humans as control over the natural environment increased. With populations growing in size, other human groups became the largest adversary and hunting tools and domesticated animals became increasingly also weapons of war, conquest and suppression. More peaceful ways of exchange intensified too: trade.

One of the most immediate and important aspects of the natural environment is the provision of food. Food is the human-environment interaction par excellence. As the environment with its geography, climate, water, soils and vegetation is the key factor in the supply of food, the focus on the emergence and forms of agriculture is natural in the present context. There are still many open questions (Messerli 2000;Walker 1993; Gunn 2000; Diamond 1997). Are there favourite sites for human societies to evolve – ecotones, interlocks of environmental zones such as hill-plain areas or river basins at medium elevation? Which factors made certain environments feasible and attractive for human habitation? How did agricultural cropping, animal herding and the use of trees for wood and fodder start – as a response to deteriorating conditions for a gathering-hunting way of life or simply, in some locations, as a more rewarding alternative strategy, or even both? Did the first farmers clearing ‘pristine’ forests, for crops, wood and fodder, result in more widespread environmental change?

Introduction

The previous chapters have given various illustrations of how socio-natural systems evolved within their environment as they developed collective and individual habits, techniques, rituals and more elaborate ways of communication and organization. These developments are sequences of processes of exploitation and adaptation. Simmons (1989) lists elements and stages of these multi-faceted interaction processes: domestication, simplification, diversification and conservation. Goudsblom (1996) stresses the differentiation in behaviour and power between people and animals and between people; increasing numbers and concentrations of people; and the specialization, organization and stratification with its interdependencies. In investigating aspects higher on the scale of complexity, there is kind of a trade-off: the distant past is badly observable but fairly simple, the near past is better communicated through writing and art but – if only for that very reason – more complex.As Roberts remarks: ‘Because information shrinks rapidly backwards through time, historical myopia is created…‘(Roberts 1989: 193) The scientific evidence, particularly regarding human-environment interactions, is also biased in other ways as explained in Chapter 5.

From all the pieces of the puzzle, hypotheses and theories have emerged to put the archaeological and historical findings and facts about past societies into a more universal, abstract perspective. They are buildings blocks towards a more comprehensive and satisfactory theory of socio-natural systems – this has not been constructed yet and possibly never will. It is worthwhile inspecting the building blocks, some of which come from the natural sciences and some from the social sciences. The hypotheses and theories often start off in a qualitative, descriptive form. In a next stage, attempts appear to mould them into the language of the natural sciences, mathematics. A system boundary is defined, relevant system elements are identified and models are constructed to explore dynamic interactions between these elements.

Introduction:Why the Roman Empire as a case study and from which perspective?

In the previous chapter we focused on what can be said about the interactions between peoples and their natural environment as social complexity increased. In a sense, the Roman Empire can be seen as the culmination of the development towards states and empires in antiquity. But can we add anything to the entire libraries of studies on the Roman Republic and the Empire that followed in its footsteps? Thousands of books have been written on all aspect of Roman society, culture, economics, military strategy and organization, measurements, infrastructures, political history – probably nothing can be added that has not already been said. But let us reverse the argument: because so much is known about the Roman period, it is a good one to present as an example of more general insights into complex socio-natural systems. There is a wealth of data, both written and archaeological, about the course of history – events as well as processes, politics as well as economics, but also commerce, literature, agronomy and almost any subject necessary to gain a thorough insight into what happened. These data have been known for a long time, and much time and effort has been expended in interpreting them.Moreover, it is quite exceptional to be able to study the full cycle of genesis, expansion, contraction and disappearance of such a major historical phenomenon that has dominated a large area of the world for over a thousand years. It allows an uninterrupted perspective on the dynamics concerned at all temporal scales – from the longest to the shortest. And because the Mediterranean Basin and its surroundings have been investigated for so long and in such detail, our perspective can also be developed for all spatial scales – from the city of Rome to the entire known world and beyond.

In the study of socio-natural relationships, this possibility of taking all scales into account is of particular importance. Firstly, among the many processes that play a role in the relations between society and the environment some are very long-term, and often cumulative.

Introduction

One of the great questions about human societies is how they emerged and transformed – and sometimes decayed – in the face of environmental change.With increased capabilities to use animals, store food and manage water supplies came a surplus of food exceeding the needs for bare survival. This allowed the rise of warriors and priests, administrations, palaces and temples – at least so the story goes. It has been related in previous chapters as the process of agrarianization, with many linked driving forces specific to given cultures and ecological regimes.Within each group of humans there would have been individuals with different skills and traits. Each group was confronted with different environmental opportunities and threats – and neighbours – and evolved in a continuous process of response and adaptation. Among some, the dominant trend may have been to live ‘the good life’. Among many, the increase in social complexity occurred in response to the need and wish to bring forth food, water and shelter from an exacting and unpredictable natural environment. In the process, some groups settled down and developed forms of agro-pastoralism that developed into large-scale land clearing and irrigation efforts; others never settled down remaining mobile nomads with large animal herds. Often neighbouring groups of humans became an ever greater enemy and war and migrations – as well as trade – intensified.

In previous chapters the natural environment has been described as a background against which the first steps into the second, agricultural regime were set.We now focus on the further unfolding of social complexity because it is an essential component of any meaningful understanding and interpretation of the human-environment relationship.1 Both extensive and intensive growth of human populations and their activities led to more intense and widespread interference with the environment and to the spreading and spatial concentrations of populations. The natural environment was the setting that provided the means for such an increase in complexity – or failed to do so. What was the role of the natural environment as a formative resource and as a constraining force?

Introduction

Jan Hendrik Weissenbruch (1824-1903) was one of the most distinguished Dutch painters in the 19th century. Skies, shores and landscapes were his passion, in particular the wide, wet ‘polders’ – stretches of land reclaimed from the ravaging waters in Holland. He never had to travel far because all this beauty was abundantly available around The Hague, the town where he had lived all his life. From his home he could walk to the famous collection of Dutch paintings at the Mauritshuis Museum in five minutes, and as a young man he spent many hours there contemplating and even copying the works of his 17th century idols, Johannes Vermeer and Jacob Ruisdael. But although he remained faithful to these great examples until the end of his days, his unrestrained abandonment to nature forced him to develop his own look at the world. ‘At times, nature gives me a real blow,’ he used to say. At such moments drawing and painting was easy. He jotted down his impressions in charcoal so that later, at home, he could work on them in paint. Over the years, his style changed from more meticulous rendering to a highly personal impressionism. What strikes the eye is the subtle balance between joyful and spontaneous virtuosity and compositional grandeur. In particular his monumental skies are unforgettable, with their infinite variety of blues and greys. He brought the polders to life and taught us to feel at home in this flat, green land of mud and water.

Although public recognition only came towards the end of his life, Weissenbruch was one of the most prominent members of the ‘The Hague School,’ a rather loose association of painters, which had its heyday in the years 1870-1900. Jozef Israëls, Jacob Maris and Anton Mauve were other well-known representatives. For many years, the 17th century artistic blossoming in the Low Countries had paralysed rather than stimulated painting in this country, but the The Hague School brought a revival.

Introduction

Our perception of the past has changed enormously over the course of time. Early travellers – geographers avant la lettre – have contributed to our knowledge by giving descriptions and conceptualizing what they encountered (Lacoste 1996). Much of this knowledge disappeared for long periods. With the ‘Golden Age’ discoveries of the 15th and 16th centuries and, in its wake, the reinterpretation or outright rejection of religious dogmas – for instance, the claim that the Earth was created in the year 4004 BC – Europe ushered herself, and the world, into the ‘scientification’ of the past. This process of making empirical observations, interpretations and experiments, i.e. the scientific method, led to the mechanization, then historization of the ‘European’ worldview, as expounded in Chapter 1. It has led to more efficient use of and increased control over the environment in the form of technology. It has also been applied to understanding and reconstructing the puzzle of our own past. The old sciences – history and geography – regained new vigour, new branches of science emerged – archaeology, palaeo-ecology and palaeo-climatology.

Using available methods and inference techniques, one can attempt to reconstruct ‘scientific facts’ about the past from what remains of it in the present. In this chapter, we focus on some of the methods to collect such empirical ‘facts’. The methods to look into the distant past are still evolving.We see more and realize that what was ‘seen’ in the past was itself part of that past – as is also true for the present. Our knowledge, both in the form of data and concepts, always came – and comes – through filters and ‘scientific’ data and theories are biased accordingly. The larger part of historic and palaeo-climatic research, for instance, has been done by European and North American researchers, and ultimately set by their research agendas. This has resulted in relatively more knowledge being available for certain places – and time periods – and in predominantly ‘Eurocentric’ interpretations. In addition, information about the past may have been or be distorted or withheld for (geo)political reasons. The more advanced societies have left conspicuous and lasting material remains which naturally became the main focus of archaeology.