Economic Integration, Land Policy and Biodiversity

Research Master Thesis Economics and Business

Economic Integration, Land Policy and

Biodiversity

Christian Bogmans

Research Master Thesis Economics and Business

Economic Integration, Land Policy and Biodiversity

Christian Bogmans

*

* Faculty of Economics & Business, Groningen, Groningen University

January 12th, 2008

Abstract

We analyze the effects of capital mobility on biodiversity and welfare. We discuss a simple general equilibrium model with trade in capital between two countries, North and South. Our model contains three factors of production: land, labor and capital. Land and capital are taxed. Biodiversity is introduced using a species-area curve. Positive investment from North to South is conditional on the ‘de-facto’ capital-labour ratio in the South, which depends not only on real endowments but also on policy. This simple framework serves several purposes. First, liberalization might lead to a decrease in global biodiversity if land policy is too lax. Second, strategic interaction between countries adjusting their policies in the face of liberalization depends on biodiversity specifics. If there is no overlap in species across countries, strategic interaction is irrelevant. Third, global biodiversity conservation is highest in the social optimum. A social planner would enforce a relatively stringent land policy in a country with a relatively high ecosystem productivity. The stringency of a country’s policy also increases if its species are highly valued. Finally, in a non-cooperative setting a classic underprovision result occurs. From a welfare point of view there is not enough habitat conservation worldwide.

CHAPTER 1: Capital mobility, Land-use and Biodiversity

...5

1.1 Introduction: Globalization and Biodiversity Conservation ... 5

1.3 A Simple Neoclassical Framework ... 10

1.4 Autarky ... 17

1.5 The Determinants of Trade in Capital between North and South... 20

1.6 Optimal Land Policy (Small Country Case) ... 21

1.7. Welfare Effects from Increased Mobility of Capital ... 23

CHAPTER 2. A Simple General Equilibrium Model for North-

South Trade in Capital and Local and Global

Biodiversity Conservation.

...29

2.1. Solving the Basic Model... 29

2.2. Growth of Factors and Technological Progress... 34

2.3 Capital accumulation, Investment and Biodiversity... 35

2.4 Environmental Kuznets Curve for Biodiversity and Land Regulation? ... 37

CHAPTER 3: Land Policy in General Equilibrium and Tax

Competition

...39

3.1 Interjurisdictional Competition : Cooperative Solution... 39

3.2 Interjurisdictional Competition: Non-Cooperative Solution... 43

3.3 Interjurisdictional Competition: Some extensions ... 46

3.4 Interjurisdictional Competition with Two Tax Instruments. ... 48

3.6 Non-cooperative Policy with Red and Green Public Goods... 51

3.7 Non-cooperative Policy with Red and Green Public Goods (Two instruments)52

Chapter 4: Trade in Goods and Global and Local

Biodiversity Conservation

56

4.1.Small Open Economy with Goods Trade and a Quota (Tax) on Land ... 56

4.2 Trade and the Environment... 56

4.3 Factor Endowment Hypothesis and Pollution Haven Hypothesis ... 58

4.4 Consumption and Production... 59

4.5 Autarky ... 60

4.6 Autarky & Optimal Land Policy... 61

4.7 International Trade... 62

4.9 International Trade & Optimal Land Policy ... 65

4.10 Welfare, Biodiversity and Land Policy ... 66

4.11 Social Planner... 67

Conclusions

...71

References...75

Appendix ‘Economic Integration, Land Policy and Biodiversity’...78

A.1.1 The Closed Economy...78

A.1.2.1 The Small Open Economy (with Ecosystem Services) ...79

A.1.2.2 The Small Open Economy (without Ecosystem Services) ...80

A.1.2.2 The Small Open Economy and Capital Taxation (without Ecosystem Services)...81

A.1.2.3 The Large Economy and Capital Taxation (without Ecosystem Services) ...82

A.1.4.1 Foreign direct investment with rent appropriation by the source country...84

A.2.1 Goods Trade: Comparative Statics Autarky:...87

CHAPTER 1: Capital mobility, Land-use and Biodiversity

1.1 Introduction: Globalization and Biodiversity Conservation

Since the beginning of the 1990’s an increasing number of economists, biologists and other scientists have paid attention to questions concerning biodiversity conservation and the rate of species extinction. Although difficult to predict, estimated extinction rates are far higher than historical rates. There is consensus that the root cause of this and other forms of environmental degradation is (economic) activity by humans. This has resulted in habitat loss, over-harvesting and pollution of the environment (see Polasky et al., 2004; Eppink & van den Bergh, 2006). At the same time, economic growth in some parts of the world has been higher than ever before, and the world has seen a rapid increase in trade flows and international investments. It seems that globalisation has taken a new pace at the same time when many ecosystems face growing pressure from human action.

According to economist Geoffrey Heal globalisation in itself has not lead to a decrease in biodiversity (Heal, 2002). Globalisation, as defined by the increase in mobility of factors of production, the lowering of transport costs and the increase in international trade and investment, cannot be the prime cause since degradation of the environment happens worldwide. Heal states that ‘Population growth, habitat loss and biodiversity loss are global problems, in the sense that they are occurring globally and have global consequences. But they are not problems of globalization (Heal, 2002)’.

The simple observation that habitat conversion and the resulting loss in biodiversity occur everywhere in the world is, according to Heal, proof of this statement. Worldwide the scarcity of land has been greater than ever not just due to globalization, but because of fast growing populations and enormous boosts in income per capita. These developments imply that many species, and nature in general, compete with other (economic) activities for scarce resources. In fact, when valued at the margin many biological assets offer such a low rate of return that from an economic point of view disinvestment is not irrational at this point in time (Bulte & Van Kooten, 2000). Even restrictive trade policies can be ineffective at preserving biodiversity since policies that reduce the terms of trade might also increase the domestic opportunity cost of preserving habitat areas (Schulz, 1996; Barbier & Schulz, 1997).

Nonetheless, Heal’s remarks can be criticized on several accounts. In sum we argue that globalization affects patterns of human economic activity with respect to space and time, and provides for many new opportunities that have no precedent in history. Let us elaborate more closely on these aspects of globalization.

and institutions seem to matter the most (Rodrik, 2003). Empirically, there is some evidence that increased openness to trade and factor flows, given the right set of domestic institutions, increases the rate of economic growth. For example, many have argued that export-led growth was one of the main drivers behind the East-Asian miracle (Rodrik, 2003). So by providing for new opportunities of economic growth, reductions in tariffs, quota’s, capital controls and other restrictions to trade and factor mobility have possibly fuelled a growth process that endangers global biodiversity. Theoretically, this would imply that the damaging scale effect from trade outweighs the preserving substitution and technology effects from trade (Copeland & Taylor, 2003)1_.

Second, Heal’s statement ignores the fact that reductions in transport costs have altered spatial patterns of economic activity. Both biodiversity and economic activity are not spread uniformly across space. For example, in the European Union there is often a larger discrepancy in income between different regions within a country than between various countries itself. Thus intra-country variety in income per capita is larger than inter-country variety in income per capita. Very much comparable to economics similar patterns of spatial heterogeneity exist in the biological realm. Some ecosystems such as tropical rainforests contain significantly more species than others. On a global scale, more than 80% of the world’s biodiversity is contained in 5% of the world’s land area. These areas are called ecological hotspots. Thus, spatial heterogeneity is an issue in both economics and ecology (Barbier & Rauscher, 2007; Eppink & Withagen, 2005) and the magnitude by which habitat conversion occurs is probably less relevant than the location where it takes place.

Third, globalization has affected the speed and scale by which mobile factors of production can relocate to other, more profitable regions. With less or no detachment to their original resource base, mobile factors have fewer incentives to acquire a sustainable relation with their environment. By the time the local environment is degraded, spatial adjustment drives mobile factors of production into unexplored areas. Thus, one should differentiate theories concerning renewable resource management with respect to factor mobility. A careful investigation of this aspect seems to be absent in most of the work on trade, renewable resources and biodiversity (Barbier & Bulte, 2005).

Footloose Capital and Labor Mobility

Since we will focus on the capital mobility aspect of globalization, let us take a closer look at some aspects of factor mobility that might be relevant for our analysis. One common aspect

of mobile factors of production is their identical impact on welfare. In standard models of international trade, trade in factors is a perfect substitute for trade in goods. Trade in factors or goods both unambiguously increases welfare if no market distortions are in place. A crucial difference however between the various mobile factors of production lies in the reason for immigration. Standard economic theory predicts that capital will flow to those regions where marginal returns are highest. In the absence of transport cost or adjustment costs these capital flows result in equalization of the net return to capital across regions. Labour on the other hand considers welfare as opposed to nominal or real returns. In many models of environmental policy and labour mobility it is assumed that agents care about total welfare (as opposed to the real wage), which includes some measure of the state of the environment as well (Hoel & Shapiro, 2003; Eppink & Withagen, 2005). If moving to a wealthy region implies a setback in environmental amenities that can be consumed in that region, then immigration might not be so beneficial than was initially thought. In other words, people’s immigration decisions depend on more than just profitability and might include, next to environmental considerations, also social, cultural and political motives (Benhabib, 1996).

Thus, without further restrictions on capital flows, there are clear differences between the incentives for immigration by mobile factors of production. In the context of biodiversity, one might conjecture that people like to move not only to those regions where real wages are high, but also where one can enjoy environmental amenities such as a diverse set of species. To date the literature has mainly focused on household mobility in the context of environmental pollution (a public bad). Even though pollution is transboundary, household mobility works as a disciplinary mechanism for competing jurisdictions to provide for an optimal amount of public goods. The intuition is that without adjustment costs jurisdictions will avoid hurting other regions, since this will eventually lead to higher pollution at home as well (Haavio, 2005; Hoel & Shapiro, 2003)

Factor Mobility and Environmental Pollution

As opposed to environmental pollution, the problem of providing for an optimal amount of biodiversity is characterized by various differences. First, environmental pollution is a public bad whereas biodiversity is a public good. Both ‘goods’ share the characteristic that they do not directly affect the budget of the government compared to standard public goods such as defence, education, healthcare and infrastructure. Of course, the main reason is that sub optimal outcomes with respect to biodiversity or environmental pollution are both the result of human action that fails to consider all external benefits and costs.

habitat. If land is an input of production, an increase in economic activity may lead to a decrease in the amount of land available for habitat purposes. As a result, biodiversity might decline. The severity will depend on the land intensity of the sector under consideration and the robustness of the ecosystem, the initial size of the habitat and connectivity to remaining habitat patches (Eichner&Pethig, 2006; Polasky, 2007; Tilman et al., 2005).

Third, for many polluting intensive industries various technologies are often there to counter emissions. Government policy in the form of emissions restrictions or taxation might induce firms to implement new technologies and abate where possible. To protect biodiversity on the other hand, fewer instruments are available so far. Species protection is in many cases considered to be expensive and habitat protection emerges as the best alternative (Bulte & Horan, 2003). The question then arises how to protect habitat from land conversion for agriculture or other economic activities.

Why Biodiversity and Factor Mobility?

The reasons for studying the relation between factor mobility and biodiversity are many. First, most work in the economic literature has only focused on the connection between trade and biodiversity. From an empirical point of view this is somewhat understandable since international trade and the associated invasion of alien species are among the most important causes of species decline (Polasky et al., 2004). Nevertheless, other threats to the environment and biodiversity exist in the form of urbanization, industrialization and eco-tourism. The latter activity is especially important for developing countries and is closely related to the mobility of capital. In particular, the underdevelopment theory of tourism describes the control of ecotourism resources by multinational enterprises in the developing world. For example, in Zimbabwe of the 1980’s more than 90% of eco-tourism revenues were expatriated to the parent countries. Only a small amount was reinvested in the home country causing excessive environmental degradation, among other problems related to sustainable development (Isaacs, 2000; Ziffer, 1989).

results are to be expected when comparing models with factor mobility on the one hand and trade flows on the other hand. We conjecture that in the area of environmental amenities, and biodiversity in particular, theories that are based solely on trade models are no substitutes for models that include factor mobility (For models with trade only see Smulders et al. (2004) and Polasky et al. (2004)).

Third, in a survey on trade and renewable resources, Bulte and Barbier (2005) recognize that considerations of factor mobility and interjurisdictional competition are absent. Combined models of factor mobility and trade would allow us to study more thoroughly the sometimes-conflicting incentives and challenges that governments face in the era of globalization. Besides caring about biodiversity and environmental amenities, jurisdictions compete to attract economic activity. In the literature on tax competition it is readily understood that factor mobility matters for policymakers. With parts of the tax base becoming increasingly mobile, interjurisdictional competition in for example environmental regulations might lead to situations of a ‘race to the bottom’. In addition, other problems with respect to the provision of public goods exist as well. In all these outcomes, fiscal externalities and environmental externalities prevent individual states choosing policies that are optimal from a collective point of view. A socially sub optimal outcome might prevail if governments fail to internalise these externalities.

Having said all this, few would argue that capital mobility is directly relevant for the current decline in biological diversity. However, capital mobility is an important ingredient in today’s process of globalization and is therefore a potential source of habitat destruction and environmental degradation. On a methodological level, a model of factor mobility, be it capital or labour, is the simplest model of economic integration. Since the global stock of capital is assumed to be constant, it allows us to isolate the integration effect from pure growth effects arising from factor accumulation. In that respect the model applied in this thesis is convenient, easy to use and adheres to the basic needs when thinking about the concept of economic integration.

Thus, there exist both theoretical and empirical reasons to study more closely the relation between factor mobility on the one hand, and biodiversity (conservation) on the other hand. In the next section we will focus first on the interaction between capital mobility, biodiversity and interjurisdictional competition. The question here is whether non-cooperative behaviour can still lead to situations where global biodiversity is protected, under the pressure of attracting capital that forms an important part of the tax base.

1.2. Interjurisdictional Competition, Capital Mobility and Biodiversity

Zodrow, 1986; Wilson, 1999). Biodiversity, like many environmental amenities, is an important example of such a public good. By its very nature though, it has some problems of its own that are not common to other public goods. First and foremost a social planner dealing with biodiversity has to consider budgetary implications, that is, money spent on for example habitat conservation is money that cannot be used to buy private goods. Second and uncommon to other public goods, the pursuit of other goals set by he government might directly interfere with the aim of preserving biodiversity. Here the challenge of attracting capital and fostering economic activity within the borders of its jurisdiction runs opposite to another need, society’s wish to protect biodiversity.

Most of the economics literature on biodiversity has focused on problems of economic policy in the context of goods mobility, i.e., models of trade and biodiversity. Polasky et al. (2004) and Smulders et al. (2004) consider the problem of habitat conversion, land-use and biodiversity in standard trade models. Relying on relatively simple connections between the ecological and the economic realm, such as the so-called species-area curve, they relate a region’s flow of goods with a region’s natural endowments and global biodiversity. Polasky et al. (2004) show that, under the right conditions, trade liberalization leads to specialization of production across countries as well as specialization in terms of species. Smulders et al. (2004) find that, in contrast to Brander & Taylor (1997, 1998) and Jinji (2006), trade measures might be beneficial for the environment in the resource-rich country if the effect of reduced harvesting of natural resources outweighs the effect of habitat destruction that is the result of agricultural expansion.

So far the issue of factor mobility has been neglected. In the era of globalization and for various reasons mentioned in the first chapter, one would expect the mobility of capital to be equally important to the issue of trade, especially giving the footloose aspect of polluting industries. This paper focuses on the issue of biodiversity in the presence of capital mobility. We will consider some basic challenges of habitat preservation and land policy that policy makers face in the context of a simple neoclassical framework where trade frictions are absent and production takes place under conditions of perfect competition and constant returns to scale.

1.3 A Simple Neoclassical Framework

Factor Endowments and Factor Mobility

normalize the price P of the aggregate good, P=1. The good is produced using three factors of production: land

T

_M, labour Land capital K. The letter M is a mnemonic for manufacturing, although the aggregate good can be interpreted as some appropriate aggregation of various sectors. Most of the time however, we will focus only on land and capital, and implicitly assume labour away. Our setting is somewhat similar to Rauscher (1997) and Wang (1995), who both consider the relation between capital mobility and environmental pollution.

Labour L and land T are immobile factors of production, whereas capital K is mobile across regions. Both regions are endowed with a fixed amount of land (and labour), a stock T(and L) in the North and a stock

_T

* _{(and L}*_{) in the South. Capital owned (}

0

K

) by Northern residents differs from capital employed (K) in the North. The stock of capital owned by the North and the South is defined as respectively

K

0

=

K

_M

+

K

_X and

* * *

0

K

M

K

X

K

=

+

, with

K

_M

0

,

K

_X

0

and similar restrictions for Southern variables. Capital owned by the North is either employed at home (

K

_M) or abroad (

K

_X ).

Capital owners are not mobile but capital itself is, and capital earnings are repatriated to the country of origin. The capital identity for the North and the South are respectively defined as K =K_M +K_X*and K* =K_M* +K_{X. However, in a two-country model the}

number of different net capital allocations is rather limited. Either North or South is a net capital investor. Thus, defining net investment I as I =K_X K_X*, we can classify capital employed as the difference between capital owned and net investments:

I

K

K

=

0 (1a)

I

K

K

=

*

+

0 * _(1b)

In what follows we continue by making use of these definitions for capital. Ecology and Biodiversity

The ecological part of the model consists of a concave relation between the amount of land available for habitat purposes THand the number of local species s, known as the species-area

curve:

)

(

T

_H

s

, *( *) H T s ,

s

T

>

0

,

s

TT

<

0

, 0, 0 * * _{> <} TT T s s (2)

H

T

s

=

,

s

*

=

*

T

_H* , 0 <1 (3)

The total endowment of land is fixed and is available for either production or habitat area:

H

M

T

T

T

=

+

. Of course, in reality species do not only survive within protected areas and there does not need to be a strict separation between economic activity and species preservation (See Polasky et al. (2005)). The species-area curve first appeared in the island-biogeography literature. This literature, initiated by ecologists MacArthur and Wilson (1967) tried to find explanations for the number of species in a particular community. The theory holds that through migration and extinction the equilibrium number of species in a particular community or island can be inferred from area size and distance from the mainland. Large islands are characterized by a larger biological diversity, as are islands close to the mainland. Depending on the latitude of the community a given area corresponds to a certain biological diversity. Areas close to the equator are found to be more ‘productive’ in terms of total species numbers.

When economists eventually started analyzing problems related to biodiversity they were looking for ways to approach this concept. Due to various reasons related to analytical tractability, ease of interpretation and ‘compatibility’ with traditional modeling techniques, the species-area curve with its concave relation between land area and species richness was found to be a suitable tool. Since then (environmental) economists when dealing with biological diversity and matters related to habitat destruction have used the species-area curve as an convenient way of bringing ecological properties into their models. The approach is often criticized for the way in which economists have adopted it; areas that are reduced to a certain size trough habitat destruction are interpreted as spanning the same number of species as an island or community that starts out with this size. Another objection, similar to the preceding one, states that increases in area-size yield higher biological diversity. However this ignores the important concept of irreversibility and assumes that migration is able to raise the level of biological diversity. In reality it may take considerable time to reverse damage from previous habitat destruction by producers. Having said all this, the species-area curve is an useful concept to use for economic modeling as long as one recognizes its limitations and is aware of the fact that results obtained are not directly suitable to guide policy analysis.

land taxation and capital taxation (see next section) are spent on a homogenous public good, M M k

K

t

T

t

G

=

+

. Production

Production in each country takes place under conditions of constant returns to scale and perfect competition,

Q

=

f

(

K

₀

I

,

L

,

T

_M

)

and ( * , *, * )

0 * M T L I K f

Q = + . The first and

second-order derivatives have the usual signs with diminishing returns to one input and positive cross-order derivatives:

0

>

i

f

,

f

ii

<

0

, fij >0 i, j =K,L,T

Producers take factor-prices as given. Government policy consists of either setting the tax-rate

M

t

on land or determining the quota

q

=

T

_M. In the first case the stock of land used in production

T

_M is endogenous. In addition, governments may also set a tax rate on capital,

)

(

*

K K

t

t

. The profit function of a representative Northern (Southern) producer is given by

M M K

K

I

wL

t

T

t

r

Q

+

=

(

)(

₀

)

(4a),(4b) * * * * * 0 * * *

₍

₎₍

₎

M M K

K

I

w

L

t

T

t

r

Q

+

+

=

Produces have to pay a tax for using capital inputs so capital is taxed in the country where it is employed. Under conditions of perfect competition (and no factor market distortions), all factors of production earn their marginal product. Profit maximization by producers thus leads to the following set of first-order conditions:

K K

r

t

f

=

+

,

f

_L

=

w

,

f

_T

=

t

_M (5a),(5b),(5c) * * K K r t f = + , * * w f_L = , f_T* =t_M* (6a),(6b),(6c) Using these factor-market conditions from the North and the South, we have a system of 6 equations with 4 or 6 exogenous variables ( _, *_{, ,} *_{, ,} *

K K M M t t t t L

Note that with a quota set in place, the factor market analysis is simplified since in that case we only have to consider the marginal productivity conditions for labour and capital. If the government of the North and South uses a land tax instead, the marginal conditions for land apply as well. Finally, in what follows we will sometimes differentiate between a small open economy case, taking r as given, or a large country case, with r endogenous. In the first case we have: * * * * * ) , , ( ) , , ( W _{K M K} K M K K L T t r f K L T t f < <

where both countries are assumed to be unable to affect the exogenous world interest rate

r

W. The interest rate ordering that is assumed here,

_r

_<

_r

W

_<

_r

*_{, indicates that the North would}

benefit from investing in the South, where the return on investment equals

r

W, and the South benefits equally since there is still a gap between domestic capital return and the world interest rate, _r* _rW _>0_{. In addition its other factors of production benefit as by the}

cooperative factors assumption.

In the second large country case we have the following so-called location condition, indicating that countries can manipulate the world interest rate and the location of capital by adjusting land policy or capital taxation:

* 0 * * * 0 , , ) ( , , ) ( _{M K K M K} K K I L T t r f K I LT t f + = = (7)

We will see that these two assumptions make a big difference for policymaking and biodiversity conservation, since it is through the interest rate that land policy has a big impact on lending, conservation and welfare.

Consumption

Utility of the representative consumer is linear additive in (physical) goods and the natural ‘good’, i.e., biodiversity. This quasi-linear function is assumed to be quite general and we restrain from specifying it any further, leaving open the possibility of private and public goods being complementary goods:

where in the absence of savings we have that consumption equals national disposable income2

)

(

)

,

,

(

)

(

0 0 0

I

K

t

T

t

I

r

T

L

I

K

f

I

r

r

rK

wL

C

K M M W M W

+

=

+

+

=

(9a)

and similarly for the South

)

(

)

,

,

(

)

(

* 0 * * * * * * 0 * * 0 * * * *

I

K

t

T

t

rI

T

L

I

K

f

I

r

r

K

r

L

w

C

K M M M W

+

+

=

+

+

=

(9b)

Income of the representative agent consists of net labour income and net capital rents that are earned from capital employed in production at home and abroad. Consumption C equals the national product Q (there are no savings) plus imports that equal net capital earnings. Consumption of public goods equals tax revenues from land and capital employed in production:

G

=

t

_M

T

_M

+

t

_K

(

K

₀

I

)

and

(

*

)

0 * * * *

_t

_T

_t

_K

_I

G

_{M K} M

+

+

=

.

The careful reader might note that the one-sector model featured in this section is devoid of consumer maximization. This feature seems somewhat unusual given the current state of economic science and might even be called old, out-dated or simply wrong. One defense against this criticism centers on the usual dogma that one should attack a model on its results and not its assumptions3_{. If the model gives intuitively and/or empirically correct}

predictions on economic integration and biodiversity, then principles of maximization are not needed per se4. Looking forward, we note that the last chapter provides the reader with a short introduction to the topic of international trade and biodiversity. Here consumers have the choice of consuming two different goods, where one of the goods produced is labor-intensive and the other land-intensive.

Returning to the ecological side of the mode, habitat loss (

dT

_M

=

dT

_H) negatively affects species numbers at home and abroad and the biodiversity index is increasing in local and foreign species numbers:

2 _{The formulation of disposable income here applies for both the small country and large} country case; in case r is endogenous (r=r*=rW) we find C=wL+rK0.

3 _{This idea goes back to Milton Friedman (1953). Friedman claimed that a theory should not} be tested on its assumptions but predictions.

)) ( ), ( ( ) , ( * * * H H as T T s b s s b B= = ,

T

=

T

_M

+

T

_H (10a) )) ( ), ( ( ) , ( * * * * * H H s T T As b s s b B = = , a,A>0 (10b)

0

<

=

S T T

b

s

b

,

b

T_*

=

b

S_*

s

T_*

<

0

Thus, a decrease in land available for habitat purposes means a decrease in local species numbers, thereby lowering the global biodiversity index. Furthermore, these definitions of biodiversity include the possibility of different valuation of foreign species and local species (a,A>0).

Balance of Payments (BoP)

Production at home can be used either for private consumption, export or turned costless into a public good provided by the government. Equations (6a) and (6b) emphasize that indeed consumption in the North (South) is larger (smaller) that the national product. One of the countries is an importer of physical or financial capital, whereas the other is an exporter of the aggregate good. In ergo, the flows of capital and goods are two sides of the same coin:

X G C Q= + + (11a),(11b)

X

G

C

Q

*

=

*

+

*

where North is assumed to be a net exporter of capital X and a net importer of the aggregate good:

rI

X =

Trade in goods is a necessary condition in this model without savings to ensure that total demand worldwide equals total supply. This can be relevant for a large number of cases, for example when the North is well endowed with capital (

K

>

K

*). Assuming otherwise symmetric countries (technology, population, land), the capital abundant North has a higher national income. The capital-rich North is a net investor in the capital-poor South and uses its return on investment to import goods from the South. The South also benefits from the improved allocation of capital since domestic factors of production, land and labor, earn a higher return as well.

Welfare

rI

T

L

I

K

f

I

K

t

T

t

rK

wL

C

=

+

₀

+

_{M M}

+

_K

(

₀

)

=

(

₀

,

,

_M

)

+

. Then, welfare in North is determined by utility derived from consumption, which equals the domestic product plus the rewards to net foreign investment, plus the benefits from global biodiversity

))

(

),

(

(

]

)

,

,

(

[

* * * 0

I

L

T

M

rI

b

s

T

T

M

as

T

T

M

K

f

u

V

=

+

+

Environmental policy or land-policy is determined by the land-tax tMor a restriction on the

use of land that is available for production,

T

_M

=

q

T

. A higher tax-rate represents, ceteris pauribus, a smaller conversion of habitat into productive land usage. In this way, a higher tax-rate protects local biodiversity.

1.4 Autarky

In autarky there is no trade in capital and goods (I =0). As a result, both countries can produce only with available domestic factors of production. We start out with the most simple situation where taxes on land are redistributed in a lump-sum fashion and taxes on capital are zero (

G

=

0

,

t

K

=

0

). Consumption equals the national product:

)

,

,

(

* * * 0 * M

T

L

K

f

C

=

)

,

,

(

K

₀

L

T

_M

f

C

=

where capital and labor are fixed and given. Land policy by the government determines the amount of land

T

_Mthat is available for production. The government maximizes utility subject to the consumption equation, i.e., the budget constraint, and a land endowment restriction. Substitution into the utility function gives us the following unconstrained problem:

max

[

(

,

,

)]

(

(

),

*

(

* *

))

0

L

T

M

b

s

T

T

M

as

T

T

M

K

f

u

V

=

+

M

T

leading to the following first-order condition for the North:

where the optimal quota is implicitly given by the equation above, ) , , ; , , ( * 0 M C M M T K L T a u

T = , where we have used a notation that separates variables from parameters. The solution for the optimal land tax tM

^

can be derived by rewriting the first-order condition5_: C T S T M

u

s

b

f

t

^

=

=

(13)

The optimal tax equates the marginal rate of substitution between land conservation for biodiversity purposes and consumption. Land policy in autarky is not determined by only domestic considerations. Optimal land policy in the form of a quota or tax would balance the benefits from habitat protection and utility derived from consumption, given the quota or tax set in the other country. In case a=0 or _b=s+s*_{, such that the North does not value}

Southern species or the ecosystems are characterized by full species endemism, the choice of the optimal quota is independent of the policy in the other country. In general it is possible to derive the ‘reaction curves’ of the North and South in autarky by totally differentiating the conditions for optimal land policy with respect to

T

_M and *

M

T

:

0

* * * 2 * * * * * * 2 * * * * * *

<

+

+

+

=

TT S T S S TT C T CC T T S S M M

s

b

s

b

f

u

f

u

s

s

b

a

dT

dT

(14a)

0

2 2 * * *

=

₊

₊

₊

<

TT S T SS TT C T CC T T SS M M

s

b

s

b

f

u

f

u

s

s

b

A

dT

dT

(14b)

In autarky land policy in the North and South are strategic substitutes. In a simultaneous move game each country will take land policy of the other as given. As a result, and depending on the models’ parameters, both will not set a sufficient amount of land aside for habitat purposes. This can easily be seen by pointing out that the optimal price of land

t

_M is equated to local, not social, marginal cost of habitat loss. A sub-optimal equilibrium arises with quotas (taxes) that are too generous (too low).

an increase in habitat area abroad. Figure 1 shows the reaction functions of the North and South, denoted by respectively R and

R

*:

A shift in the reaction function of the North, for example due to an increase in the labor force, shifts the equilibrium to E^. In the North more land is set aside for production and less for habitat purposes. For the South this strategy results in a shift along its reaction curve. For the South the optimal response is to counter the loss in global biodiversity by setting more land aside for habitat at home. This is a typical result that is obtained in other fields of economics as well, such as strategic interaction in a duopoly, i.e., a Cournot-Nash game (See Tirole, 1988). In this context we have applied it to land policy and biodiversity. Here each country has an incentive to lower its land tax or increase its quota in order to set more land aside for production, while at the same time expecting the other country to take the burden and to increase habitat conservation to preserve global biodiversity.

1.5 The Determinants of Trade in Capital between North and South

In the model specified we have assumed that trade in capital will equalize its return r in all countries. This assumption implies that barriers to trade are completely absent and that capital markets are fully integrated. Before working out questions with respect to optimal environmental policy in such a context, let us consider what happens when countries initially open up to trade with an exogenous world interest rate. Assuming (1) an exogenous world-price r for capital and (2) capital scarcity (abundance) in the South (North),

_K

_>

_K

*_{, this}

condition is given by * * * * * ) , , ( ) , , ( W _{K M K} K M K K L T t r f K L T t f < <

If countries are identical in terms of technology (

f

=

f

*) and labour endowments (

L

=

L

*), differentials in gross autarky prices of capital are driven by different (1) capital endowments (

_K

_K

*_{), (2) different tax rates on capital (} *

K

K t

t ) or (3) different land/environmental quotas (T_M T_M*)6_{. In autarky, changes in the marginal product of capital are driven by}

changes in land policy:

df

_K

=

f

_KT

dT

_M and df_K* = f_KT*dT_M*. Thus more land raises the productivity of capital, a standard result in models with cooperative factors of production.

Proposition 1. A country attracts foreign capital, i.e., it is capital-poor, if its autarky stock of capital is relatively small and/or its land quota is relatively high and/or if the tax on land-use is relatively low.

As a simple exercise consider what happens to consumption, biodiversity and welfare in autarky when the South changes its land quota T_M*. Since biodiversity is an (imperfect) global good the North is affected by this change as well:

0 * * = S T < M s b a dT dV

Production and therefore consumption in the South rise due to the extra input of land in production. Biodiversity declines since habitat area is converted into ‘productive land’. Welfare unambiguously declines in the North, because global biodiversity decreases. For the South we find ambiguous welfare effects because utility derived from consumption increases, whereas non-use value from biodiversity declines. The overall effect will depend on the marginal rate of substitution between consumption and biodiversity, u_C* / *, and the land policy that is in place.

1.6 Optimal Land Policy (Small Country Case)

In the previous section we showed that the welfare effects for the country imposing a change in land policy are ambiguous. Unless an optimal policy is in place, a marginal change in land usage for production, either as result of a change in the quota or a surge in investment from abroad that raises the demand for land, does not automatically lead to welfare gains (

dV

/

dT

M

>

0

). The optimal autarky tax rate on land relates the benefits of more land usage in the form of higher consumption against the damage to biodiversity. Thus, as we showed in the previous section there exists an obvious trade-off between consumption and biodiversity conservation.

Next, we consider how changes in some important model parameters affect land policy. To determine how the marginal valuation of biodiversity and ecological productivity affect the optimal land policy (tax or quota), we totally differentiate the first-order condition for optimal land policy of the South with respect to *

M

T

,

T

_M, , * _and *_:

(

)

(

S T S S T

)

S S T S S T T M T S M TT S T S S F CC TT C dT s s b A d s s b d s b s b d s b dT s b s b f u f u * * * * * * * 2 * * * * * * * * * * * * * 2 * * * * 2 * * * * + + + = + + +

The following comparative statics results can be derived from this equation:

0

* * * 2 * * * * 2 * * * * * * * *

+

+

+

=

TT S T S S F CC TT C T S S M

s

b

s

b

f

u

f

u

s

s

b

d

dT

(15c)

Countries with a relatively large marginal valuation of biodiversity have a greater incentive to implement a more strict land policy in the from of a high tax or quota. Somewhat more complicated is the effect of a country’s ecosystem productivity on its environmental policy. There are two conflicting forces. First, there is a ‘positive’ effect from increased ecosystem productivity on the stock of land used in production. Since the biodiversity index is assumed to be concave,

b

_S*S*

<

0

, the positive effects of increases in carrying capacity eventually ‘die out’. Thus, there comes a point where greater environmental capacity needs to be ‘traded’ for more productive land use (income effect). Second, there is a negative effect from an increase in carrying capacity. A higher carrying capacity increases the returns from ‘existing’ habitat area ( *

_>

0

T

s

), inducing the country to even increase the stock of land devoted to habitat. We can summarize this in the following proposition.

Proposition 2. If a country attracts foreign direct investment for other reasons than those concerning endowments, it does so because of relative lax environmental policy. In turn, lax environmental policy itself can be rooted in relatively small preferences for biodiversity and/or relatively large ecosystem productivity. However, a relatively ecosystem productivity may actually lead to a more stringent land policy if the biodiversity index is not very concave and/or the existing habitat is relatively large.

From this we see that a country may chose to set an tax rate on land that induces specialization in nature if the ‘substitution effect’ of ecosystem productivity is larger than the ‘income effect’ from ecosystem productivity.

The last comparative static result, which summarizes the marginal change in the optimal quota from a marginal change in foreign ecosystem productivity, is unambiguously positive. Thus, if some exogenous shock negatively affects carrying capacity, thereby reducing biodiversity abroad, the other country will, ceteris pauribus, definitely reduce land usage for domestic production. However, two important remarks are in order here. First of all, it is not always realistic to have an exogenous shock that only affects ecosystem productivity in one country. Probably more plausible would be a shock, for example as the result of climate change, causing a reduction in worldwide ecosystem productivity, _{d = d} * _<0

of climate change. The overall optimal changes in land policy can then be given by differentiating both first-order conditions for optimal land policy:

* * * 2 * * * * 2 * * * * * * 2 * * * * * * * *

₍

₎

TT S T S S F CC TT C T S S T S S T S M

s

b

s

b

f

u

f

u

s

s

b

s

b

s

b

d

dT

+

+

+

+

+

=

(16a) TT S T SS T CC TT C T SS T SS T S M

s

b

s

b

f

u

f

u

s

s

b

s

b

s

b

d

dT

+

+

+

+

+

=

₂ *₂ * 2

_/

₎

(

(16b)

where we have neglected the impact of foreign ecosystem degradation on land policy at home (

_dT

/

_d

*

₌

_dT

*

/

_d

₌

0

M

M ), which is in line with the Nash-equilibrium concept, that is, the

country under consideration takes the policy in the other country as given,

₌

*

₌

0

M

M

dT

dT

.

The derivatives show that, besides the standard income and substitution effects we discussed in the preceding section, there now is a an extra term that makes the overall effect more likely to be positive. This implies that compared to a shock that only affects local ecosystems, a global shock is more likely to reduce land usage for production. Note that these effects are deduced under full information of the shock,_{d = d} * _<0_{. With full information, the local}

government knows that not only local but also foreign ecosystems are degraded which asks for a more severe reduction (smaller increase) in land usage for production.

1.7. Welfare Effects from Increased Mobility of Capital

The welfare effects from increased openness to capital markets depend partially on the environmental policy or land policy that is in place. In the previous section we showed that land policy that is either to protective or too lax could cause sub-optimal outcomes in the face of increased openness. Welfare can even decline if habitat area is excessively converted into productive land area, causing extinction of a large number of local species. If a strict policy, that is, an enforceable quota ( *

₌

0

M

dT

) is in place the welfare effects are unambiguously positive:

(

*

)

0

* * * *

>

=

=

u

f

r

dI

dC

u

dI

dV

K C C

f

K

>

r

* ₍₁₇₎

in turn increases the demand for land. As a consequence, habitat area is converted into land area for production (agriculture, manufacturing etc.) leading to

) ( ) ( * * * * * * * * dI dT s b dI dT s Ab dI dT f r f u dI dV _M T S M T S M T K C + + = (18)

showing that under a tax-policy the total effect on welfare from a marginal increase in investment depends on factor market interactions. Differentiation of the factor market condition for land in North and South we get */ ₌ * / * _>0

TT TK M dI f f dT and

0

/

/

=

TK TT

<

M

dI

f

f

dT

. Substitution of these derivatives into (eq.12) leads again to ambiguous welfare effects in the South under conditions of increased capital mobility:

) ( ) ( ? * * * * * 0 * * * * * 4 4 4 4 3 4 4 4 4 2 1 4 4 4 3 4 4 4 2 1 TT TK T S TT TK T S TT TK T K C f f s b f f s Ab f f f r f u dI dV ₌ >

A marginal increase in investment raises capital productivity in the South and increases the demand for land, which all leads to an increase in consumption that positively affects welfare in the capital-scarce South. This is the first term. The second term is ambiguous and represents the global change in biodiversity. In the country where capital is leaving more land becomes available for habitat purposes, whereas in the South production is intensified and local species are under pressure due to loss of habitat. The overall effect on welfare from a marginal increase in investment is not clear.

Proposition 3. Under a tax-on-land regime increased openness to foreign direct investment leads to an increase in land as a factor of production, an increase in consumption and a decrease (increase) in local (foreign) biodiversity. If the tax on land is relatively small and/or foreign biodiversity is not highly valued (small A) and/or the foreign species-area curve is very concave, then welfare in the capital-poor country may decline.

Note that with an optimal tax in place, _{* *} *

* * T S C M

b

s

u

0

)

(

* * *

>

=

TT TK T S K C

f

f

s

Ab

r

f

u

dI

dV

under the optimal tax more land is set aside for production in the country where investment takes place, but only to an extent where the marginal damage from local biodiversity loss equals the marginal increase in utility from consumption.

Again, in the absence of an optimal policy trade in capital is not guaranteed to increase welfare in the capital-poor region. The derivation of the optimal tax was relatively easy, but in practice it requires a high degree of knowledge of the world’s ecosystems and the marginal valuation of global biodiversity and consumption. In view of these somewhat unrealistic information requirements let us consider a country that aims to adjust its policy in response to increased openness to capital. We assume that the initial policy in place is not optimal, being either too strict or too lax. A reduction in barriers to capital mobility might lead to economic and environmental changes that, in the long-run, provide for more information on the structure of ecological and economic systems. In time, a country opening its borders might adjust its policy in response to capital inflows. To determine the South’s (North’s) optimal response to a change in its capital stock recall the first-order conditions for optimal land policy:

0 = = M M C M dT db dT dC u dT dV , _*

0

* * * * * * *

=

=

M M C M

dT

db

dT

dC

u

dT

dV

,

Totally differentiate these first-order conditions for optimal land-policy with respect to

T

_M,

*

M

T

and I to obtain the following ‘reaction’ functions for the North and South:

) ( ) ( 2 * * * * * * 2 * * * * * * * * * * * T S S TT S T CC TT C M T T S S TK C K T CC M s b s b f u f u dI dT s s b f u r f f u dI dT + + + + + = (19a)

)

(

)

(

2 2 * * * T SS TT S T CC TT C M T T SS TK C K T CC M

s

b

s

b

f

u

f

u

dI

dT

s

s

b

f

u

r

f

f

u

dI

dT

+

+

+

+

=

(19b)

these derivatives implicitly contain the reaction functions ( *)

capital. Simplifying notation such that

dT

_M

/

dI

₌

[

A

D

(

dT

_M*

/

dI

)]

/

C

_and

* *

*

*

_/

_dI

_[

_A

_D

₍

_dT

_/

_dI

_)]

_/

_C

dT

_M

=

_M , we can obtain the ‘general equilibrium’ effects of a marginal change in investment on land policy by simple substitution:

*

* * *

DD

CC

DA

A

C

dI

dT

_M

₌

+

*

)

(

* * * *

DD

CC

A

D

CA

dI

dT

_M

₌

+

since C<0, _C* _<0_{, A>0, D 0, D}* _{0 but}

_A

* _{ambiguous, we find that both}

derivatives are ambiguous; only in case of full species endemism (_{D= D}* ₌0_{) do we find}

some clear-cut results, that is, the North reduces its land usage in production.

If we neglect for a moment the other country’s reaction to an increase in capital mobility7_{, then we observe that for the North the optimal response is to reduce its land quota,}

that is, tighten its land policy (

dT

_M

/

dI

<

0

). This means that the initial positive (negative) effect from capital outflow on local biodiversity (production) is strengthened furthermore. For the South, the optimal response is ambiguous and will depend on the direct ( * *

TK C f u ) and indirect effects (

_u

_f

*

(

_f

*

_r

)

K T

CC ) of extra capital on the productivity of land. The indirect

effect implies that the use of more land in production becomes less useful due to diminishing returns to consumption in utility.

When a country recognizes the impact of its own land policy on the other region, then the sign of this strategic effect is determined by the sign of

b

_S*S, the cross partial of the global biodiversity index. This derivative considers the effect from a marginal increase in Northern species numbers on the marginal increase in biodiversity from Southern species numbers, and vice versa. We consider two extremes (See Polasky et al.(2004) and Barbier&Rauscher (2007)):

• High Species Endemism,

B

=

b

(

s

,

s

*

)

=

s

+

s

*

Ecosystems in the North and South may be completely different and give home to a vast amount of species that are all country specific. Under this condition we observe that habitat destruction, which is the result of capital-led growth in industrial or agricultural activity, may lead to the extinction of a number of species that are unique to the booming region and for which no ‘substitute’ exists in other regions. High endemism lowers the probability that an increase in local species numbers makes an additional specie in the other region redundant. Thus, the cross partial is negative but small in terms of absolute value. For the extreme case of absolute species endemism, the partial is exactly zero,

b

S_*S

=

b

SS_*

=

0

.

• High Redundancy,

_B

₌

_b

(

_s

,

_s

*

)

₌

max{

_s

,

_s

*

}

At the other side of the spectrum we may find a situation of high redundancy. Now both regions have very similar ecosystems and contain a set of local species that is found in the other region as well. Taken to the extreme, global biodiversity is just the maximum of species numbers’ living in one of the two regions. Habitat destruction in one region does not necessarily lead to global extinction of some species. Here we find that the cross partial is negative and large in absolute value. Under high redundancy an increase in local habitat area and species numbers most probably makes an additional specie in the other region obsolete,

0

* *S

=

SS

<

S

b

b

.

Making use of these various forms of the biodiversity index, we can formulate the following propositions.

Proposition 4. Without taking into account the other region’s change in land policy, a resource-rich country should reduce its land quota in response to increased openness. For the country that is relatively poor in capital the optimal response in land-policy is ambiguous.

Proposition 5. If countries act strategically, that is, acknowledging conjectural variations, then the specific functional form of the global biodiversity index determines the optimal response. Under strict species endemism,

b

S_*S

=

b

SS_*

=

0

, the aforementioned strategic effect completely disappears. In case of redundancy of local and foreign species, the optimal response of the capital-poor country to a change in the other’s regions land policy is negative. For the resource-rich country the optimal response is negative as well.

) ( / / 2 * * * * 2 * * * * * * * T S S TT S T CC TT C T T S S M M s b s b f u f u s s b dI dT dI dT + + + = (20a)

)

(

/

/

2 2 * * * T SS TT S T CC TT C T T SS M M

s

b

s

b

f

u

f

u

s

s

b

dI

dT

dI

dT

+

+

+

=

(20b)

In response to the North’s initial reduction in its land quota, the South has an extra incentive to loosen its land policy even further. First, there is the initial reaction to the inflow of capital that induces the South to increase (decrease) its quota (tax) op to the point where the marginal loss of local biodiversity equals the marginal gain in utility from consumption. Second, there is the increase in habitat area in the North that induces the South to loosen its land policy further. This positive externality lies at the root of the further conversion of habitat area in the South; the social gains of habitat protection in the North are larger than the private gains, and the South is willing to ‘substitute’ some of these gains for extra consumption. The North has a similar incentive, knowing that further habitat destruction will increase the return to its investment and increases habitat area at home to make up for the foreign loses in biodiversity.

CHAPTER 2. A Simple General Equilibrium Model for

North-South Trade in Capital and Local and Global Biodiversity

Conservation.

2.1. Solving the Basic Model.

The first chapter of this thesis focused on economic integration and biodiversity conservation in the context of a small open economy. Though we mentioned the possibility of extending our analysis to a two-country model with endogenous interest rate determination, we did not explicitly formulate and solve such a model. In this chapter we do formulate such a model and solve it explicitly by using specific functional forms.

To solve the model we use Cobb-Douglas production functions in both countries. We assume technology across countries is identical, but make no specific assumptions about relative factor endowments, preferences and land policy. The model is summarized in Table 1 below.

Table 1: Summary of the two-country model with capital mobility

North South Production Functions: M

T

I

K

L

Q

=

(

₀

)

* * 0 * * _{( )} M T I K L Q = + (T1.1)

Factor Demand Functions (Labor and Land):