ESSAYS ON PUBLIC FINANCE AND ENVIRONMENTAL ECONOMICS IN NAMIBIA

Transcription

1 Michael N. Humavindu: Essays on Public Finance and Envionmental Economics in Namibia ESSAYS ON PUBLIC FINANCE AND ENVIRONMENTAL ECONOMICS IN NAMIBIA by Michael N. Humavindu ISBN: ISSN: Umeå Economic Studies No. 705 UMEÅ UNIVERSITY 2007 ISSN

2 Copyight Michael N. Humavindu Depatment of Economics Umeå Univesity 2007 ISSN: ISBN: Akitektkopia, Umeå 2007

3 Essays on public finance and envionmental economics in Namibia by Michael N. Humavindu

4

5 Abstact This thesis compises two papes exploing aspects of public finance and envionmental economics in Namibia. Pape [I] estimates the shadow pices of capital, labou and foeign exchange fo the Namibian economy. The esults suggest that the shadow pice of capital fo Namibia is 8%. The economic costs of Namibian labou, as a shae of financial costs, ae 32% fo uban semi- and unskilled labou, and 54% fo ual semi- and unskilled labou. The economic cost of foeign labou as a shae of financial costs is 59%. The estimated ange fo the shadow exchange ate facto is between 7% and 14% fo the Namibian economy. Pape [II] studies the deteminants of popety pices in the township aeas of Windhoek, the capital of Namibia. The study eveals that popeties located close to an envionmental hazad, such as a gabage dump, sell at consideable discounts. On the othe hand popeties located nea an envionmentally favouable location, such as eceational open space, sell at a pemium. These esults povide evidence of the impotance of envionmental quality in lowe income popety makets in developing counties. It is theefoe impotant fo Namibian uban plannes to incopoate envionmental quality within the planning famewok fo lowe income aeas.

6

7 Acknowledgements I would like to thank the following people and institutions who, infomally o fomally, have contibuted to this wok and have helped me duing my studies: Jespe Stage, my supeviso, mento and co-autho on one of the papes in this thesis. Wods of gatitude could neve be enough to thank you fo what you have done, and continue doing, in helping me become a competent economist woking fo the development of my county. Hopefully one day, you faith in me will be justified. Kal-Gustaf Löfgen, David Potts, Kenneth Backlund, Jon Banes and Tomas Sjögen, fo comments on ealie vesions of Pape I (the shadow pices pape). The Jan Wallande and Tom Hedelius Foundation, fo the eseach gant that made all of this possible. Finally, many thanks go to my fiends and family, who ae too numeous to mention. You constant encouagement is geatly appeciated. The wok is dedicated to my gandmothe, Kauipondua Humavindu, and my son, Tjaiove Mbajooka Humavindu. Umeå, 19 Apil, 2007 Michael Nokokue Humavindu

8

9 This thesis compises an intoduction and the following two self-contained papes: [I] Humavindu, M.N. (2007). Estimating national economic paametes fo Namibia. [II] Humavindu, M.N. and Stage, J. (2003). Hedonic picing in Windhoek townships. Envionment and Development Economics, 8: Repinted with pemission fom the Cambidge Univesity Pess.

10

11 Intoduction and Summay Intoduction and Summay 1 Intoduction This thesis consist of two self contained papes, analysing public finance and envionmental economics aspects of Namibia. Pape [1] estimates shadow pices fo the Namibian economy fo use in social cost benefit analysis. Pape [2] analyses the deteminants of popety pices in a low income aea of Namibia s capital, Windhoek. The focus is on the implicit valuation of envionmental advantages and disadvantages in low income popety picing makets. The common thead linking the two papes is that an application of compehensive cost-benefit analysis in Namibian developmental pojects would enhance decision-making. The application of consistent shadow picing in Namibian pojects would be a useful tool fo national plannes, even at an uban level. Economic tools fo poject appaisal in developing counties have been well established since the ealy 1970s. The appopiate appaisal of public investment pojects undelines the need to detemine the social value of costs and benefits accuing fom these investments. In developing counties especially, social values may divege fom maket pices and values, depending on the level of distotions. These distotions may be caused by maket impefections as a esult of both govenment inteventions in poduct and facto makets, and stuctual disequilibia in labou makets. As a esult of the distotions, maket pices ae uneliable indicatos of the eal net woth of goods and sevices (Adhikai, 1986). A case in point is the divegence fom its tue value of the official pice of foeign exchange, due to govenment s decision to devalue the cuency. Official tade policy, such as the adoption of vaious taiff and non-taiff tade baies, may also lead to a distoted maket value of foeign exchange. The esult is a distotion in the domestic pice of all tadeables, and also of non-tadables which use tadeables in thei poduction. In labou makets, the equilibium wage may be highe than the efficiency wage, as a esult of minimum wage laws and union bagaining pesence. In capital makets, the maket inteest ate may divege fom the maginal poductivity of capital. 1

12 Intoduction and Summay In poject appaisal, modifications to maket values ae thus essential. A modification is detemined by estimating a set of national paametes and convesion factos. These paametes ae temed shadow pices. Convesion factos give the atio between the pice to be used in evaluating the poject (the shadow pice), and the pice used to evaluate its cost (the maket pice). In the valuation of inputs used in poduction, the inheent assumption is that the pice of any input should epesent the oppotunity cost of that input. The oppotunity cost eflects the value of output fogone on one poject when used on anothe. Shadow pices ae thus useful when the maket pice fo an input o output is unavailable. Labou, fo example, is an impotant input in many investment pojects, and should theefoe be valued at its economic cost. Shadow pices ae a cucial link between the maco level and the poject level of economic planning, and an impotant component of the oveall pocess of development planning in developing counties. Only when the costs and benefits of all potential pojects ae valued at thei shadow pices, may those pojects that most efficiently use scace esouces be selected. Following this stategy would allow a developing county to maximise the potential net economic benefits accuing fom its investments, theeby impoving its potential to pusue boade social, political and othe non-economic objectives (Saebeck, 1989). In geneal national paametes to be estimated fo economic analysis ae divided into five categoies (Potts et al, 1998): pimay factos, taded goods, non-taded goods, aveage estimates and the discount ate. Pimay factos elate to diffeent categoies of labou, the value of domestic esouces and foeign exchange. Taded goods ae goods fo which the economic cost o benefit deived fom thei use is detemined by thei intenational pices. Shadow pice estimation is essential whee thee is a significant diffeence between the bode pice and the local maket pice. Deiving the shadow pice is also a necessity in situations whee a benefit is likely to featue pominently as an input o output fo a numbe of pojects. Non-taded goods ae items that, by thei natue, cannot be taded acoss bodes, o may not be economically viable fo tade. The estimation of a shadow pice is pompted by a 2

13 Intoduction and Summay situation whee thee is a significant diffeence between the local maket pice of a esouce and its economic value, o whee, as fo taded goods, a benefit is likely to featue pominently as an input o output fo a numbe of pojects. Aveage estimates elate to sectos whee cost data do not allow futhe beakdown. The most impotant of such estimates, the standad convesion facto, descibes the value of a unit of domestic esouces in tems of a unit of foeign exchange. The standad convesion facto, in the case of aveage estimates, is deived indiectly though convesion factos fo taded goods. Discount ates quantify the effect of time on a poject s cost and benefit values. Classical shadow picing estimation would involve deiving a geneal equilibium economic optimisation model with the following specific featues (UNIDO, 2000): an objective function descibing the effects of the use and geneation of esouces on a measue of economic value such as the goss domestic poduct (GDP) constaints on the use of esouces (technological coefficients fo each economic activity and a limit fo the esouce as a whole) non-zeo constaints fo the value of esouces and non-negativity constaints fo esouces. The shadow pice is then the effect on the value of the objective function esulting fom an incease o decease by one unit in the availability of a scace esouce. Howeve the estimation of shadow pices in this way is faught with complexities and numeous constaints, and is infeasible in pactice (Little & Milees, 1974). This has led to the adoption of 'second best' appoaches to shadow pice estimations. These methods wee developed in the late 1960s and ealy 1970s by the United National Industial Development Oganization (UNIDO, 1972, 1978, and 1980), and by Little & Milees (1974) and Squie & van de Tak (1975). The depatue point between these two appoaches is the choice of unit of account. 3

14 Intoduction and Summay In essence, the UNIDO appoach uses a domestic esouce as the unit of account and estimates the scacity value of foeign exchange using a shadow exchange ate. This pocedue is descibed as the use of a domestic pice numeaie (Potts, 2002). The second appoach, developed by Little & Milees and Squie & van de Tak, uses the unit of foeign exchange (expessed in local cuency units) as the numeaie. This method is descibed as the use of a wold pice numeaie. A thid appoach is based on the pemise that all shadow pices ae intedependent because thei value depends on the value of inputs fom othe sectos (Potts, 2002). These intedependencies ae accounted fo though convesion factos that ae deived by solving a seies of simultaneous equations using an input-output appoach. This appoach thus takes into account all the sectoal inteelationships; it is called the semi-input-output analysis (SIO), and is useful fo non-taded sectos whee the output fom each secto may appea as inputs into othes. In estimating shadow pices, the choice of methodology is pimaily detemined by the natue and extent of available data. Readily available data was a constaining facto in this study, which theefoe attempts initially to estimate the thee pimay factos, capital, labou and the exchange ate.. Futue extensions to this wok would examine the possibility of employing the SIO analysis whee data pemits. Despite the appaent impotance of shadow picing fo a developing county such as Namibia, no set of official national paametes exists, no has any attempt been made to estimate them. The county s development path is guided by five-yealy development plans (National Development Plans). These national plans stess the impotance of investment/development pojects to alleviate chonic unemployment, low industialisation, povety and income inequality. Unde such cicumstances, it is vital that maket signals povide an adequate guide fo investment planning and poject appaisal. Thee is an appaent need fo a consistent set of pices which eflect the esouce costs and social benefits of a poposed couse of action. High unemployment (36%), uneven income distibution and an economy that expots most of its capital ae all stong motivations fo the estimation of a set of national paametes. Recently eleased national guidelines 4

15 Intoduction and Summay ( Guidelines fo Pepaing the Thid National Development Plan (NDP3): 2007/ /12 ) e-asset the impotance of investment/development pojects fo Namibia s economic development. In esponse, Pape [1] is the fist fomal attempt to estimate the shadow pices of capital, labou and foeign exchange fo the Namibian economy. The estimation is based on data epesenting national aveages. Shadow pices based on national data aveages have to be distinguished fom sectoal, egional o poject-specific paametes (Saebeck, 1989). Poject-specific paametes should be estimated fo each individual poject, because the oppotunity costs of the esouces used o poduced may diffe fom poject to poject, due to the specific chaacteistics of each poject. This can be applied, fo example, to aspects of uban planning. The economic value of an uban housing poject fo lowe income esidents may be highe, if it is located nea envionmentally beneficial featues (such as paks) and public amenities (such as schools, taxi anks), than one located nea envionmental hazads, o fa fom public amenities. Pape [2] is an application of the hedonic picing methodology (Rosen, 1974) to study the deteminants of popety pices in a low income aea of Namibia s capital, Windhoek. The methodology uses popety pices to estimate buyes implicit valuation of a popety s attibutes (such as access to public sevices, poximity to envionmentally beneficial o detimental featues) when tading takes place. Economic decisions ae made evey day with egad to the use of all types of esouces. In allocative decisions, pices act as a signaling device. The use of some esouces, howeve, does not fall unde the ambit of the maket. These ae efeed to as non-maketed esouces; they ae not bought and sold in the maket place, and ae consequently unpiced. The missing maket phenomenon theefoe ceates economic development biases against decisions that benefit the envionment, and in favo of decisions that ham the envionment. The valuation of envionmental assets and sevices would undeline thei economic impotance and make a case fo thei consevation. The incopoation of envionmental costs and benefits falls within the ambit of non-maket valuation in the envionmental 5

16 Intoduction and Summay economics discipline. Non-maket valuation is a measue of the willingness to pay fo the value of unpiced envionmental goods and sevices. Geneally, fo non-maketable items (those that cannot be sold o bought), two goups of valuation methods ae employed. 1 The fist method is the evealed pefeence appoach, in which consume behaviou towads envionmental goods is analysed and values ae infeed. Peoples pefeences ae evealed by thei choices. Thee ae two widely used evealed pefeence methods. The fist is the tavel cost method, in which the costs of tavel ae used as the substitute fo what people would be willing to pay fo the oppotunity of using a public envionmental benefit such as a pak o open aea. The tavel cost method is used to estimate the value of eceational benefits geneated by ecosystems. It assumes that the value of the site (o its eceational sevices) is eflected in the amount people ae willing to pay to get thee. It is efeed to as a evealed pefeence method, because it uses actual behaviou and choices to infe values. The application of the tavel cost method has been extended into the andom utility modelling methodology. The andom utility appoach is the most complicated of the tavel cost appoaches. It is also the appoach of choice, because it allows fo much flexibility in calculating benefits, making it the best appoach fo estimating the benefits of specific chaacteistics, o quality changes, of a site, athe than of the site as a whole. It is also the most appopiate appoach when thee ae many substitute sites. The second impotant evealed pefeence method is hedonic picing, which postulates that the pice of a commodity is elated to its chaacteistics. Theefoe, vaiations in demand fo a commodity (such as a house) can be statistically elated to its attibutes (e.g. local ai quality, amenities). The hedonic picing method is used to estimate the value of envionmental amenities that affect the pices of maketed goods. Most applications use esidential housing pices to estimate the value of envionmental amenities. The method is based on the assumption that people value the chaacteistics of a commodity, o the sevices it povides, athe than the commodity itself. Pices will thus eflect the value of a set of chaacteistics, including envionmental chaacteistics, that people conside impotant when puchasing the commodity. 1 See OECD (1995) and Hufschmidt et al (1983). 6

17 Intoduction and Summay Revealed pefeence appoaches ae useful whee consumes ae making eal decisions about envionmental goods and sevices. Howeve such appoaches ae of little assistance when consume behaviou towads envionmental goods cannot be obseved. The solution then is to apply what is temed as the stated pefeences appoach. This is the second goup of nonmaket valuation methodologies. The appoach ests on the simple pemise of putting hypothetical questions to consumes. The most widely used method of hypothetical pefeence appoaches is the contingent valuation technique, in which suveys ae used to ask espondents how much they would be willing to pay to ensue a given envionmental impovement. The contingent valuation method (CVM) is used to estimate economic values fo all kinds of ecosystems and envionmental sevices. The method has geat flexibility, allowing fo the valuation of a wide vaiety of non-maket goods and sevices than is possible with any othe non-maket valuation technique. It can be used to estimate both use and non-use values, and it is the most widely used method fo estimating the latte. It is also the most contovesial of the non-maket valuation methods. The fact that contingent valuation is based on what people say they would do, as opposed to what people ae obseved to do, is the souce of its geatest stength and its geatest weakness. Contingent valuation is one of the few ways to assign monetay values to non-use benefits attached to the envionment. Citicisms of envionmental valuation ae ooted in both technical and theoetical issues. Technical poblems ae elated to data limitations and inheent biases in valuation techniques. Such poblems might lead to inaccuate values. The second categoy of citicisms has to do with theoetical, philosophical and ethical objections to envionmental valuation. Notable objections cite the issue of distibution acoss and between geneations, the poblem of using human pefeences as the basis of envionmental value, and the difficulty of aiving at values when the effects of envionmental factos ae ievesible o when envionmental isks ae unquantifiable. Despite these citicisms, valuation methods have, in ecent decades, eached a consideable degee of sophistication, and thee has also been notable consensus on thei use (Secetaiat of the Convention on Biological Divesity, 2007). The only diffeences that emain on these 7

18 Intoduction and Summay valuation methods ae on the level of teminology and of classification. As a esult, valuation methods ae inceasingly applied, not only in developed counties, but also in developing counties. Rietbegen-McCacken & Abaza (2000) explained that: [U]p to ecently, thee was consideable skepticism, paticulaly among intenational development oganizations and developing county govenments (as end uses of the valuation esults) about the possibilities of using valuation methods outside the elatively esouce-ich and data-ich envionments of developed counties. It was geneally felt that developing counties and counties with economies in tansition pesented too many difficulties (including a scacity of statistical infomation; the pesence of pice distotions o undeveloped makets; and in some cases lagely illiteate communities) to allow valuation methods to poduce meaningful esults. Howeve, ove the last five to ten yeas a gowing body of evidence has emeged to efute these claims. Rietbegen-McCacken & Abaza (2000) povided a numbe of case studies of valuation studies in Afica, Asia, Latin Ameica and Cental and Easten Euope. Some of these case studies also dealt with biodivesity esouces and functions as the elated ecosystem sevices. IUCN (1998) guidelines fo potected aeas manages on economic values of potected aeas povided summaies on a few valuation studies in developing counties. FAO (2001) conducted a suvey on the use of contingent valuation studies in developing counties. In Namibia the wok by Humavindu & Masiembu (2001) and Humavindu (2002) wee two pioneeing examples. Humavindu & Masiembu (2001) estimated the diect value of Windhoek s eceational assets using contingent valuation and tavel cost methods. The study found that eceational assets wee peceived as impotant by Windhoek s inhabitants and that they wished to have them peseved. Humavindu (2002) pesented an analysis of valuation studies addessing natue-based touism, and established that touists wee willing to pay moe to visit and peseve Namibia s touism assets. 8

19 Intoduction and Summay 2 Summay of the papes Pape [1] Estimating national economic paametes fo Namibia In the fist pape of this thesis, shadow pices of capital, labou and foeign exchange fo the Namibian economy ae estimated. Although the use of shadow pices is essential fo sound developmental planning, the application of shadow picing in Namibia has been limited o vitually non-existent. The inteest in deiving Namibian shadow pices aises fom both pactical and academic points of views. In pactical tems, the ecognition of the need fo lage scale investments to dive economic gowth pompts the need to apply shadow pices, so as to ensue optimal allocation of scace esouces. Fom an academic point of view, the Namibian economy exhibits special featues that suppot the need to estimate national paametes. A highly uneven income distibution, a lage infomal economy, and minimum wages in cetain sectos all validate the necessity of estimating shadow wage ates. Unlike most othe developing counties, Namibia is a net capital expote. Although the economy has high domestic savings, the lack of domestic investment oppotunities leads to capital outflow amounting to 10% of GDP annually. The shadow pice of capital can then be easonably expected to be low. Finally, Namibia s membeship of the Southen Afican Customs Union (SACU) and the Common Monetay Aea (CMA) might affect the estimation of the shadow pice of foeign exchange in Namibia. SACU goups togethe Botswana, Lesotho, Namibia, Swaziland (bette known as BLNS counties) and South Afica, and applies a common extenal taiff. The SACU Ageement has ecently been enegotiated, with key elements evised and given a new focus, in the light of the need to allow BLNS counties geate say in the detemination and administation of SACU taiffs. The CMA compises SACU counties, excluding Botswana, and is a monetay aea with a centalised monetay policy aimed at achieving geate financial stability fo the egion. The monetay policy is contolled by South Afica, and all othe CMA cuencies ae pegged to the South Afican Rand. In pinciple thee should be two Shadow Exchange Rates (SERs): one fo convetible cuency extenal to the Common Monetay Aea, and one fo Rand based cuencies, which 9

20 Intoduction and Summay would have a shadow exchange ate of 1, since thee ae no tade estictions between the CMA counties. The SER calculated in this pape is applicable to tansactions with counties outside the CMA, but not to the foeign content of goods puchased fom South Afica. The esults suggest that the economic oppotunity cost of capital is 8% in Namibia. The economic costs of Namibian labou, as a shae of financial costs, ae 32% fo uban semiand unskilled labou, and 54% fo ual semi- and unskilled labou. The economic cost of foeign labou, as a shae of financial costs, is 59%. The estimated shadow exchange ate facto lies between 7% and 14% fo the Namibian economy. Pape [2] Hedonic picing in Windhoek townships The second pape attempts to detemine whethe popety pices in seveal townships of Namibia s capital, Windhoek, ae affected eithe by positive o by negative attibutes, and applies the hedonic picing method. Hedonic picing, as peviously stated, involves the implicit pice of attibutes o chaacteistics of a commodity athe than the pice of the commodity itself. Hedonic picing models ae used to infe the demand fo attibutes of envionmental quality, though the analysis of maketed goods whose value depends, in pat, on these attibutes. The methodology is geneally applied fo the valuation of envionmental goods, popety and wate, and the implicit pice of attibutes and chaacteistics of the maketed goods. The geneal assumptions of such a model ae that all the goods o sevices bought to the maket must be clealy visible, and that popety values and the implicit pice of attibutes o chaacteistics should be teated as a single maket. Unde these assumptions, the pice of any esidence can be descibed as a function of the envionmental, stuctual, and neighbohood chaacteistics of the esidence location. The Hedonic model gives a ealistic estimation of envionmental benefits and values, as model estimates ae based on maket infomation. In this pape, we use popety sales data obtained fom the City of Windhoek Municipality, and apply the hedonic picing model. Ou findings ae that, apat fom housing quality, access to the cental business distict, access to maketplaces, and access to tanspotation, envionmental quality also has a lage impact on popety pices. Popeties located close to 10

21 Intoduction and Summay a gabage dump sell at consideable discounts, while popeties located close to a combined consevation and eceation aea sell at pemium pices. The esults thus suggest that the hedonic picing method can be usefully applied when studying townships in developing counties, and that this can claify and emphasise the impotance of envionmental factos, which ae othewise fequently neglected in town planning fo township settlements. Refeences Adhikai, R National Economic Paametes fo Nepal. Occasional pape no 9. Poject planning cente fo developing counties. Univesity of Badfod. FAO (2001): Applications of the Contingent Valuation Method in Developing Counties: A suvey. FAO Economic and Social Development Pape 146, Food and Agicultue Oganization of the United Nations, Rome. Habege, A. (1978). 'On the Use of Distibutional Weights in Social Cost Benefit Analysis', Jounal of Political Economy, 86 (2). Hufschmidt, M.M., D.E. James, A.D. Meiste, B.T. Bowe, and J.A. Dixon (1983): Envionment, Natual Systems, and Development: An Economic Valuation Guide. Johns Hopkins Univesity Pess, Baltimoe, MD. Humavindu, M. N. (2002): Economics without Makets: Policy Infeences fom Natue-Based Touism Studies in Namibia. Diectoate of Envionmental Affais (DEA) eseach Discussion pape 49, Ministy of Envionment and Touism, Windhoek, Namibia. Humavindu, M.N. and S. Masiembu (2001), The Economic Value of Uban Receation in Namibia: Results of a Pilot Study at Avis and Goeangab Dams, Diectoate of Envionmental Affais Woking Pape No. 7, Diectoate of Envionmental Affais, Ministy of Envionment and Touism, Windhoek. 11

22 Intoduction and Summay IUCN (1998): Economic Values of Potected Aeas. Guidelines fo Potected Aea Manages. Wold Commission on Potected Aeas (WCPA), Best Pactice Potected Aea Guidelines Seies No. 2. Little, I. and Milees, J. (1974). Poject Appaisal and Planning fo Developing Counties. London: Heinemann. National Planning Commision Guidelines fo Pepaing the Thid National Development Plan (NDP3): 2007/ /12. Govenment of the Republic of Namibia Windhoek, Namibia. OECD The Economic Appaisal of Envionmental Pojects and Policies: A pactical guide. Pais: OECD Publications. POTTS, D Poject Planning and Analysis fo Development. Lynne Rienne Publishes. London, UK. Potts, D., Weiss, J., Beyene, M., Guta, F., Kinfu, H. and Assegid, R. (1998). National Economic Paametes and Economic Analysis fo the Public Investment Pogamme in Ethiopia. Addis Ababa: Ministy of Economic Development and Co-opeation. Rietbegen-McCacken, J. and H. Abaza (2000): Envionmental Valuation. A Woldwide Compendium of Case Studies. Eathscan, London. Rosen, S. (1974), Hedonic Pices and Implicit Makets: Poduct Diffeentiation in Pue Competition, Jounal of Political Economy 82: Saebeck, R National Economic Paametes fo Botswana. Reseach Monogaph 1. Development and Poject Planning Cente. Univesity of Badfod. Secetaiat of the Convention on Biological Divesity, (2007). An Exploation of Tools and Methodologies fo Valuation of Biodivesity and Biodivesity Resouces and Functions, Technical Seies no. 28, Monteal, Canada, 71 pages. 12

23 Intoduction and Summay Squie, L. and van de Tak, H. (1975). Economic Analysis of Pojects. Johns Hopkins Univesity Pess. UNIDO Guidelines fo Poject Evaluation. United Nations. UNIDO Guide to Pactical Poject Appaisal. United Nations. UNIDO Pactical Appaisal of Industial Pojects: Application of Social Cost Benefit Analysis in Pakistan. United Nations. Winpenny, J.T Values fo the Envionment: A Guide to Economic Appaisal. London: ODI. 13

24 Intoduction and Summay 14

25 ESTIMATING NATIONAL ECONOMIC PARAMETERS FOR NAMIBIA by MICHAEL N. HUMAVINDU 1 ABSTRACT This pape estimates the national economic paametes to be used fo poject appaisal in Namibia. The shadow pices of capital, labou and foeign exchange ae deived. The esults suggest that the economic oppotunity cost of capital is 8%. The economic costs of Namibian labou, calculated as a shae of financial costs, ae 32% fo uban, and 54% fo ual semi- and unskilled labou. The economic cost of foeign labou, calculated as a shae of financial costs, is 59%. The estimated shadow exchange ate facto lies between 7% and 14% fo the Namibian economy. 1 The autho wishes to acknowledge a eseach gant fom the Jan Wallande and Tom Hedelius Foundation. The autho would like to thank Jespe Stage, Kal-Gustaf Löfgen, David Potts, Kenneth Backlund, Jon Banes and Tomas Sjögen fo comments on ealie vesions of the pape. The usual disclaime applies. Fo coespondence please the autho at m_humavindu@hotmail.com.

26

27 Estimating National Economic Paametes fo Namibia 1. INTRODUCTION In this pape I estimate the shadow pices of capital, labou and foeign exchange fo the Namibian economy. Shadow pices ae defined as the oppotunity costs of inputs and outputs consumed o poduced by a poject (Potts, 2002). The value that a esouce could have geneated elsewhee in the economy is lost if that esouce is moved to a paticula poject. Shadow pices ae theefoe calculated to take into account the tue oppotunity costs of esouces, inputs and any extenal factos esulting fom a developing pogamme o poject. In many makets, especially in developing counties, financial o maket values diffe fom thei eal economic values due to distotions bought about by impefect o undedeveloped makets, govenment potection policies and othe extenal factos (Behman, 1986). The geatest distotions ae associated with unskilled labou, the cost of foeign exchange, and the cost of financial capital. Shadow picing is then used to coect fo these distotions and value esouces, and to appoximate thei actual value. The use of unadjusted maket pices fo labou and capital could lead to an undeestimation of the eal costs of capital intensive pojects, pomoting these at the expense of the socially less costly labou intensive pojects. High levels of effective taiff potection, togethe with impot quotas and ove-valued exchange ates, disciminate against the agicultual secto in favou of the impot-substituting manufactuing secto. Such distoted domestic poduct pices not only eflect inaccuately the eal tems of tade between, fo example, agicultue and industy, but also tend to favou uppe income goups dispopotionately in elation to lowe income goups. It is thus essential to estimate shadow pices when evaluating a poject in tems of an economic analysis. Poject evaluation aims to induce allocation efficiency in the use of a county s esouces by applying cost-benefit analysis (Campbell & Bown, 2003). Despite its cucial ole in sound developmental planning, the application of shadow picing in Namibia has been limited o vitually non-existent. This is unfotunate, as Namibia's 3

28 Estimating National Economic Paametes fo Namibia development stategy, encapsulated in the five-yealy National Development Plans and in Vision 2030, undepins the impotance of development/investment pogammes in addessing the challenges of povety, high unemployment, inequality and low industialisation. 2 The launch of a national development bank in 2004, to fund long-tem infastuctue pojects, futhe emphasises the impotance of undestanding the economic costs and benefits of the pojects it suppots. Potential lage scale pojects, such as the development of Kudu Gas Fields, Tansfontie Touism Paks and othe infastuctue pojects, would need to be assessed on both financial and economic gounds. Pactically applied in the economic analyses of Namibia's development pojects, shadow picing has the potential both to ensue the optimal allocation of the county s scace esouces, and to help the county to attain the tagets set out in its development stategy. Estimating shadow pices fo the Namibian economy is also inteesting fom an academic point of view. Namibia has special featues that ae not commonly found among othe developing counties. Namibia s Goss Domestic Poduct (GDP) pe capita at maket exchange ates (2005) of US$ 3,100 is elatively high fo a developing county. Howeve, accoding to the Wold Bank's Wold Development Indicatos 2006, Namibia has the wold's highest Gini index (74.3, compaed to South Afica's 57.8 and Botswana s 63). 3 This implies an uneven income distibution, which povides futhe incentives fo estimating shadow wage ates. The economy has an unusually lage sevice secto fo a developing county (aound 58.7% of GDP). Independence in 1990 bought consideable changes to the economy's extenal and intenal migation pattens, especially in elation to the labou maket. Accoding to Fayne & Pendleton (2001) intenal migation and ubanisation in Namibia is gowing apidly, diven to a lage extent by employment oppotunities in uban centes. The population of Windhoek, the capital city, gew on aveage at an annual ate of 5.4% in the 1990s. No substantial eseach has been caied out in Namibia on eithe the scale o the 2 Shadow pices ae consistently used in the Ministy of Envionment and Touism (MET), but these ae based on educated guesswok athe than eal estimates. This famewok assume the economic oppotunity cost of capital at 8%, an adjustment (up by 6%) to the value of tadable goods to eflect excess demand fo foeign exchange, and an adjustment (down by 65%) to unskilled labou costs, to eflect unemployment (Banes, 1994). 3 The Gini index is a measue of the degee of income inequality. 4

29 Estimating National Economic Paametes fo Namibia possible consequences of skills emigation, but both skilled and unskilled Namibian migatoy labou, deployed to South Afica (SA) and to othe neighbouing counties, is vey limited, accoding to peliminay analyses by Fayne & Pendleton (2001, 2002) and the Migation Dialogue fo Southen Afica (MIDSA, 2006). Namibia s net migation ate is estimated at 0.47 fo evey 1000 pesons. In addition, unlike othe developing counties, Namibia is a net capital expote. The economy has high domestic savings, but these flow out in the most pat to SA, in seach of highe etuns. The lack of domestic investment oppotunities is cited as one eason fo the pesistent capital outflows (Fitch Ratings, 2005), which amount to 10% of GDP annually, and continue unabated. The linkages to SA ae not only esticted to Namibia's capital outflows. The two economies ae membes of egional goupings such as the Southen Afican Customs Union (SACU), the Southen Afican Development Community (SADC) and the Common Monetay Aea (CMA). Namibia's cuency is pegged to the South Afican Rand; 82% of the county s total impots ae fom SA, and aound 26% of the county s total expots ae to SA. The South Afican economy, the egional economic powehouse, is some 30 times the size of the Namibian economy. Recent calculations by Habege et al (2003), Kuo et al (2003) and Bicak et al (2004) poduced estimates of shadow pices fo labou, capital and foeign exchange in the South Afican Economy. Since South Afica and Namibia shae simila histoical and political ties and a cuent close economic elationship, it would be inteesting to compae the esults of this wok with those fom the South Afican studies. Howeve given Namibia's special featues, atypical fo a developing economy, it can easonably be expected that the national paamete estimates fo the two economies will be diffeent. This study is the fist fomal attempt to estimate shadow pices fo the Namibian economy. The pape is stuctued as follows. Section 2 (following) discusses economic featues petinent to the estimation of Namibian shadow pices. Section 3 is a teatment of appoaches to shadow picing as well as the methodology employed. Section 4 descibes the 5

30 Estimating National Economic Paametes fo Namibia data employed in the analysis and the assumptions undelying each estimate. Section 5 pesents the esults and Section 6 concludes. 2. FEATURES OF THE ECONOMY PERTINENT TO ESTIMATION OF SHADOW PRICES (a) Dynamics of Capital Makets The Namibian financial makets exhibit special featues that affect the estimation of the shadow pice of capital. Limited investment oppotunities in domestic financial makets have led to sizable outflows of Namibian savings into the liquid and elatively developed South Afican makets. Membeship of the CMA also allows fo fee capital outflows, and equies Namibia to confom to South Afican exchange contol pactices fo counties outside the CMA. These outflows aveaged aound N$1.8 billion pe yea between 1990 and 1994, and have acceleated to about N$2.4 billion pe yea since 1995, diven by net outflows in both potfolio and othe investments. The Namibian economy is pimaily esouce-based, and thus has some investments that ae highly pofitable, fuelled by esouce ents (economic pofits obtained by utilising natual esouces). These ents, existing because the natual esouces ae scace, can be an impotant souce of development finance; counties like Botswana and Malaysia have successfully leveaged natual esouces this way. Howeve, in sectos that do not have esouce ents, the maginal poduct of capital appeas to dop shaply, as many funds ae invested outside of Namibia. The Namibian authoities have followed a two-ponged stategy to stem capital outflows: egulatoy contols, to estict capital outflows, and development of domestic makets, to povide institutional investos with assets denominated in Namibian dollas. The latte stategy is still in its infancy. In the mid 1990s Namibian authoities aised egulatoy equiements fo both the pension fund and the insuance industies (Regulations 28 and 15 espectively), such that 35% of the assets unde thei management had to be domestic assets (up fom an ealie 10%). This action did contibute to the gowth of the Namibian Stock 6

31 Estimating National Economic Paametes fo Namibia Exchange, due to an incease in dual listings of South Afican companies. Howeve these investments in dual-listed companies wee unable to contain capital outflows. As a esult the govenment has poposed futhe changes to tighten the domestic assets equiements: among othe new poposals, a 5% minimum fo unlisted investments and a 10% maximum on dual listed shaes. The impact of the egulatoy contols is significant. The total demand fo Namibian investment instuments is estimated at N$ 23 billion whilst the total supply of Namibian investment instuments is estimated at N$ 17 billion. Thee is thus a N$ 6 billion excess supply of funds, which is cuently channeled to dual-listed shaes on the Namibian Stock Exchange. (b) Labou Maket Dynamics The Namibian labou maket is govened by a policy famewok which includes a labou act, a social secuity act, an employment policy, an affimative action bill and incentives fo investment and taining. On balance, howeve, unemployment and undeemployment emain high. Accoding to the latest Labou Foce Suvey of 2004, unemployment is estimated at 36%. Accoding to the Bank of Namibia s annual epot (2004), an estimated 15% of the employed population ae undeemployed. Motinga & Tutalife (2006) indicate that Namibia ceated a mee 22,000 fomal jobs between 1991 and Unemployment dispopotionately affects the Namibian youth and the unskilled. The duation of unemployment is longe in ual aeas, and vaies between six months and 2 yeas. Thee is also evidence of wage inequality between the skilled and the unskilled. Motinga & Mohammed (2002) calculated that an aveage unskilled peson eans 3% of the wages and salaies of top management, and less than 50% of what an aveage skilled peson eans. Westegaad-Nielsen et al (2003) confim the huge wage diffeentials between skilled and unskilled labou. Although thee is no fomal minimum wage legislation, industy-specific wage ageements fo the constuction, agicultue and secuity industies do contain minimum wages. Thee is also a lage infomal economy employing at least 133,000 people, 7

32 Estimating National Economic Paametes fo Namibia 64% of whom ae young people. Remuneation in this secto is vey low, and thee ae no benefits such as social secuity and medical aid. The pesence of a lage infomal economy and minimum wages lead to Namibian wages being set highe than the maginal poductivity of labou. The case fo such an economic adjustment is made by the impefections in the labou makets. The infomal economy, which consists of lage numbes of small scale businesses, can easonably be assumed to be a secto with maket cleaing wages. In the fomal secto howeve, the pesence of minimum wages, collective bagaining and potential efficiency wage issues, leads to wages above the maket cleaing levels found in the infomal economy. Due to these distotions, a potion of the 36% unemployed would pefe fomal jobs, but cannot get them. (c) Issues in Estimation of Foeign Exchange Pemium Namibia s paticipation in the Southen Afican Customs Union (SACU) aises the question of whethe such paticipation affects the estimation of the shadow pice of foeign exchange (SER). SACU goups Namibia, Botswana, Lesotho, Swaziland and SA togethe and applies a common extenal taiff. All customs and excise duties collected by the five membes of SACU ae pooled into a Common Revenue Pool (CRP), and e-distibuted among them accoding to a Revenue Shaing Fomula (RSF). The shaing of the evenue fom customs duties is detemined on the basis of each county's pecentage shae of total inta-sacu impots, excluding e-expots, and not on the basis of SACU impots fom the est of the wold (Flattes & Sten, 2005 and Kik & Sten 2005). The majoity of Namibian impots (82%) ae fom SA, which inceases Namibia's shae of the evenues fom the SACU system (due to the RSF inta-sacu impots ule). Namibian impots fom outside SACU (the emaining 18% of the county s total impots) ae subject to SACU taiffs, but geneate vey little exta SACU evenue fo Namibia because taiff evenues ae paid into the SACU system, fom which funds Namibia eceives only a small potion. Most Namibian expots ae to counties outside SACU, and theefoe do not affect the county s evenue shae fom SACU. 8

33 Estimating National Economic Paametes fo Namibia Thus since the Namibian SACU evenue is effectively independent of Namibian out-of- SACU impots, it can be agued that SACU eceipts ae not elevant to the detemination of the shadow exchange ate, since they ae essentially inte-govenmental tansfes, and do not diectly affect the elationship between pices of taded goods at wold pices and thei domestic pices. Moeove the SACU evenue pool is now gadually declining due to continuing tade negotiations at the multilateal and egional levels. In elation to the customs union, thee is a question as to how to teat impots within the union. Namibia is pat of the CMA, which includes Namibia, Swaziland, Lesotho and South Afica. 4 With the exception of Botswana, the CMA has moe o less the same membe counties as SACU; thee should thus be two shadow exchange ates (SERs) one fo convetible cuency extenal to the CMA, and one fo Rand-based cuencies which would have an exchange ate of 1, since thee ae no tade estictions between the CMA counties. The SER to be calculated in this wok is theefoe applicable to tansactions with counties outside of the CMA, but not to the foeign content of goods puchased fom South Afica. In pinciple one would expect the SER to be simila in elation to extenal economies fo all membes of the CMA, because they ae all using the same taiff stuctue; howeve thee might be some vaiation due to diffeences in the stuctue of impots. 3. ANALYTICAL FRAMEWORK (METHODOLOGY) This section descibes the analytical famewok within which shadow pices in Namibia ae estimated. Geneally thee ae two appoaches to shadow picing, hinging on assumptions egading the existence of maket distotions (Medalla, 1982). The fist appoach may be genealised as an attempt to estimate shadow pices associated with a fist-best optimum. If maket and shadow pices divege due to policy failues, then the appopiate shadow pices would be the equilibium pices that would pevail if the distotions wee emoved. If, howeve, the divegence is caused by failues of maket, athe than policy, then non- 4 The CMA is descibed as an aea of coodination of the monetay and exchange ate policies of its membes unde the Multilateal Monetay Ageement of Unde the CMA, the Namibian cuency is linked one-toone to the South Afican and, which is also legal tende. The CMA also guaantees fee capital flows among membe counties, as well as access fo Namibian govenment and financial institutions to South Afica s financial makets. See Tjiongo (1995) and Vollan (2000). 9

34 Estimating National Economic Paametes fo Namibia optimality is in essence caused by the absence of fist-best coective measues. The studies by Tinbegen (1958) and Bacha & Taylo (1971) in the case of shadow picing of foeign exchange ae associated with this fist appoach, which, fo pimay factos such as capital and labou, is not yet feasible due to inadequate techniques and data (Medalla, 1982). The second appoach teats pesent distotions as given, and assumes that they might pesist in the long tem (Medalla, 1982). Shadow picing is then a poblem of deiving the dual solutions to the optimisation of welfae, while the distotions ae teated as constaints. Unde this appoach the poblem of optimisation is usually not fomally specified, but athe foms the conceptual famewok fo shadow picing ules. The esulting shadow pices ae efeed to as second-best shadow pices, epesenting the social costs and benefits of inputs at the second-best optimum. This appoach is associated with the wok of Little & Milees (1969), Habege (1972) and Dasgupta et al (1972). In this pape we follow Habege s (1972) appoach, fo two easons. Fistly, accoding to Khan (1979), this is the coect method of estimating the shadow discount ate whee the maginal social value is not equal to the maginal social cost of funds at the maket equilibium, due to the pesence of vaious distotions. Secondly, and moe impotantly, this appoach will allow fo the compaison of these esults with those of Habege et al (2003), Kuo et al (2003) and Bicak et al (2004) fo South Afica. (a) Economic Oppotunity Cost of Capital (EOCK) Habege & Jenkins (2002) and Baeix (2003) mention fou diffeent appoaches to the choice of a discount ate fo use in economic cost benefit analysis. Howeve, only one appoach (the EOCK Appoach) is suitable fo a small open economy like that of Namibia. The othe thee appoaches ae maed by vaious constaints. 5 5 It is not the pupose of this pape to delve into a lengthy exposition of the vaious appoaches to defining the social ate of etun. One appoach suggests that both pivate and public investments should be discounted at a ate equal to the maginal poductivity of capital in the pivate secto. A second appoach ecommends the use of an accounting ate of inteest which is the estimated maginal etun fom public secto pojects, given the fixed amount of investment funds available to the govenment. A thid appoach ecommends that the benefits and costs of pojects should be discounted by the social ate of time pefeence fo consumption, but only afte costs have been adjusted by the shadow pice of investment to eflect the fact that fogone pivate investment has a highe social etun than pesent consumption. The eade is advised to consult Baeix (2003), who povides an excellent summay. 10

35 Estimating National Economic Paametes fo Namibia The EOCK appoach postulates that in a small open and developing economy, like Namibia, thee ae thee altenative souces fo public funds. The fist souce comes fom individual saves, who take esouces that would othewise have been spent on pivate consumption, and instead contibute to an incease in domestic savings. The second souce is fom additional foeign capital inflows. The thid souce is fom those esouces that would othewise have been invested in othe investment activities that have been eithe displaced o postponed by the poject s extaction of funds fom the capital maket (Habege & Jenkins, 2002). Based on these thee altenative souces of public funds, the economic cost of capital can be estimated as a weighted aveage of the ate of time pefeence applicable to additional savings, the maginal cost of additional foeign capital inflows, and the ate of etun on displaced o postponed investments. In geneal, vaious distotions ae associated with each of these thee altenative souces of funds. Theoetically the social ate of etun may be defined by applying national accounting pinciples. In an open economy, eal income may diffe fom eal poduct because of the sevicing of national debt. Let us assume that s is the aveage inteest ate on the stock of foeign debt (D). Then income Y is given by: Y = q - s.d (1) whee q is the eal poduct. If we then conside a new public poject: ΔΥ = δ. ΔI + ρ. ΔI i. ΔD (2) g p f whee q= δ. Δ I + ρ. ΔI g p, q is the pemanent change in eal poduct, I g is the new public poject, δ is the ate of etun of the poject, I p is the change in pivate investment caused by the new poject ( I p <0), ρ is the maginal ate of etun that the postponed investment would have geneated, i f is the maginal cost of additional foeign boowing and D epesents the change in the extenal debt stock. 11