Chapter Twelve. Economic Opportunity Cost of Labour

Transcription

1 Chapter Twelve 12.1 Introduction The concept of economic opportunity cot i derived from the recognition that when reource are ued for one project, opportunitie to ue thee reource are acrificed elewhere. Typically, when worker are hired by a project, they are giving up one et of market and nonmarket activitie for an alternative et. The economic opportunity cot of labour (EOCL) i the value to the economy of the et of activitie given up by the worker, including the non-market cot (or benefit) aociated with the change in employment (Harberger, 1980). When determining the EOCL, it i important to remember that labour i not a homogeneou input. It i perhap the mot divere factor of production in any economy. In thi chapter, we will examine how the EOCL i etimated in an economy that contain market for many different type of occupation, with variation by region and by quality of employment opportunitie (e.g., pleaant, unpleaant, permanent, temporary, etc.) that affect the EOCL for a project. The focu i primarily on the condition and ditortion in the labour market; the dicuion at thi point will not concern the potential impact that employment of dometic labour might have on the market for aving or foreign exchange. 1 1 In the evaluation of the EOCL, we do not take into account the potential impact on national aving of change in the amount of income received by labour. Thi deciion i baed on three obervation. Firt, mot of the labour hired by a particular project would have been employed elewhere in the abence 335

2 A labour externality ( LE i) i created for any project when the economic opportunity cot of labour ( EOCL i) differ from the wage rate ( Wp i) paid to the labour by the project. Thi externality can be expreed for a pecific type of labour (i) a: of the project. Second, the overall level of national aving i fundamentally determined by macroeconomic and public ector budgetary condition. Third, the level of uncertainty urrounding the quantitative etimate of the ize of the ditortion attributed to aving, and the impact on national aving of labour receiving more or le income from a project, warrant coniderable caution. If, however, a particular project i deemed to have a meaurable impact on aving, and there i an externality aociated with thi impact, then the value of thi externality hould be included in the evaluation of the economic net preent value (NV) of that project. In a imilar manner, we do not take into conideration the indirect effect on ditorted market, uch a foreign exchange, that are due to the movement of labour from other activitie to the project. If the quantitative impact of the indirect effect that occur through the foreign exchange market or any other ditorted market i known, the value of thi externality hould be included in the evaluation of the economic benefit and cot of the project. 336

3 LE i= Wp i EOCL i (12.1) When LE i i poitive, the financial cot of labour will be greater than it economic cot and vice vera. A thi analyi will how, the magnitude of thi externality i a function of more variable than imply the rate of unemployment in the relevant labour market for thi cla of worker. It will alo depend on other ditortion in the labour market, uch a taxe, unemployment inurance, and protected labour market egment. Furthermore, it will be affected by the quality of the job created. The magnitude of thi externality i one factor that caue the economic performance of a project to diverge from it expected financial outcome Alternative Approache to Etimating the In etimating the EOCL, two alternative tarting point for analyi may be choen: a) the value of the marginal product of labour forgone; and b) the upply price of labour. It hould be noted that calculating the EOCL uing either method will theoretically produce the ame reult. However, thee two approache have different data requirement, level of computational complexity, and, hence, different degree of operational uefulne Value of Marginal roduct of Labour Forgone The value of marginal product of labour forgone for labour hired by a project i determined by tarting with the gro-of-tax alternative wage ( W a ) that the labour hired for that project would have earned in it abence. In mot cae there will, at any future point in time, be an etimated ditribution of the labour activitie in the preence of the project and an alternative ditribution in it abence. The difference between thee two allocation will uually um to zero, epecially if leiure and involuntary unemployment are counted among the relevant 337

4 activitie. Thi mean that the net reduction in labour allocated to other activitie mut add up to the amount of employment provided by the project. If one work trictly with forgone marginal product, the EOCL for the project would imply be the weighted um of the forgone marginal product of labour of all different type ourced from the variou activitie. Thi method i not well adapted for reflecting difference in the underlying working and living condition that do not directly reduce output elewhere in the economy (Little and Mirrlee, 1974). Hitorically, ome economit have argued that the value of the marginal product of unkilled agricultural worker in developing countrie wa zero becaue it wa believed that there wa a large urplu of labour in the countryide. However, empirical tudie of ubitence farmer have demontrated that their labour doe have a poitive marginal value, both in farming and in a variety of other productive activitie. Uing the aumption that the value of the marginal product forgone i zero when hiring unemployed worker, thi approach lead to an underetimation of the EOCL, and the etimate doe not reflect the true economic cot of the project uing the labour Supply rice of Labour An alternative method that i baed on the upply price of labour i more traightforward and eaier to ue under a wide variety of condition (Harberger, 1971). The tarting point i the market wage (the upply price) required to attract ufficient people of the required kill level to work on the project. The upply price of labour to a project i the minimum wage rate that the project need to pay in order to obtain ufficient upplie of labour with the appropriate kill. That wage will account for the worker preference regarding the location, working condition, or any other factor that affect the deirability of working for the project. For example, if a very high local market wage i required to attract killed labour to a project where the living condition are bad, 2 For a ummary of thi debate, ee Marglin (1979, pp ). 338

5 that wage already include both the value of the forgone wage and the compenation for the economic cot inflicted by the relatively bad living condition. Of coure, the upply price hould be adjuted further to account for other ditortion, uch a taxe, to arrive at the EOCL. But unlike the marginal product forgone approach, where one mut meaure both of thee component eparately, the local upply price directly capture in a combined package the wage and non-wage cot of employing labour on the project. In practical term, the upply price of labour can be determined by etablihing the minimum wage the project mut pay in order to attract an adequate number of applicant to work on the project with an acceptable turnover rate. Thi can often be done by informally urveying worker near the location of the project or uing a more formal aement of the prevailing wage in that activity. To tet whether the wage rate being paid by a project i the minimum upply price, one hould compare the number of application by qualified people with the number of poition available. If the number of acceptable application per job available i very high, and the turnover rate for the project i abnormally low, it i likely that the wage rate paid by the project i above the minimum upply price. However, if the ratio of qualified applicant to poition available i low, thi indicate a fairly tight labour market, and the turnover rate i high for the type of kill required. In thi cae it can be aumed that the project wage i cloe to the minimum upply price of labour. Once the minimum upply price of labour ha been determined, the EOCL i calculated by adjuting that value to account for relevant ditortion (uch a income taxe or ubidie). Care mut be taken at thi point to enure that all of the market ditortion that drive a wedge between the upply price and the EOCL are properly accounted for when etimating the EOCL for the project. The evaluation of a number of thee ditortion i covered in the following ection of thi chapter. To compare thee two method of calculating the EOCL, let u conider the example of unkilled farm worker who have decided to move from their job cutting ugar cane (c) to work on a new project in a more pleaant place (o), harveting orange. The tarting point for calculating the EOCL uing the marginal product forgone approach would be the alternative wage on the ugar 339

6 cane plantation farm ( W c ), while the upply price approach would begin with the market wage for working in the orange grove ( W o ). It i aumed that thee worker do not pay income taxe or face any other ignificant ditortion in their labour market. Other factor, however, could influence their deciion to relocate to the new project. For example, the more pleaant climate of the orange-growing region might tranlate into a reduced cot of living (C), which would allow the worker to maintain the ame level of well-being with lower wage. Another factor might be a preference (S) on the part of worker to work in a more pleaant region. For the purpoe of thi example, it i aumed that value of the wage and the other factor are a follow: W o = $15.00 per day, W c = $20.00 per day, C o = $3.00 per day, C c = $6.00 per day, S o = $2.00 per day (value of the preference for the warmer region) Uing the marginal product approach, the EOCL can be calculated for the new project a follow: EOCL = prior wage change in cot of living worker preference = W c ( C c C o ) S o = $20 ($6 $3) $2 = $15.00 per day Uing the upply price approach, it i poible to arrive at the ame value directly becaue it i known that the market wage neceary to induce the worker to move to the new project in the orange-growing region ( W o ) already account for the cot of living difference ( C c C o ) and the worker regional preference for the better climate ( S o ). Therefore, the EOCL i imply equal to the market wage in the region where the new job i located: 340

7 EOCL = W o = $15.00 per day Thi highly implified example demontrate that both method for calculating the EOCL hould produce the ame reult. However, in mot circumtance it i difficult to place value upon complex factor uch a cot of living differential and worker regional preference. Uncertaintie in the value of thoe factor make the marginal product forgone approach cumberome to ue when information i carce. Conequently, the traightforward upply price approach i uually an eaier way to determine the EOCL Structure of Analyi in the Labour Market The analyi of the EOCL preented here i tructured around five et of factor, which are primary determinant of the cot of labour to a project. Labour price can vary greatly from one project to the next, o the following claification are ued to help to identify which of the determinant may have an effect on the labour cot of the project being evaluated: Type of labour (killed v. unkilled) Regional variation and dometic migration Type of job (permanent v. temporary) International migration Type of labour market (protected v. unprotected) Firt, an analytical ditinction i made between different kill and occupation. Claifying worker into relevant occupational categorie i eential becaue of the enormou heterogeneity of the labour factor. In general, the lower the kill, the greater the likely homogeneity of labour within the kill or occupational category. Etimating the economic opportunity cot of unkilled labour i alo made more traightforward by the frequent abence of ditortion uch a taxation and unemployment inurance in that part of the labour market. The killed 341

8 labour market, on the other hand, diplay much greater heterogeneity and i frequently ubject to multiple ditortion, which mut be identified and accounted for in the etimation of the EOCL. Second, regional migration induced by difference in wage, cot of living, and acce to conumer good and amenitie alo affect the EOCL for a project. Regional wage differential are a key conideration in the labour market, where a rie in project employment in an urban etting ha a it counterpart reduction in employment in rural area that are traditional ource of migration. In uch cae, ditortion in the economy related to that migration mut be accounted for when etimating the EOCL. Third, it may alo be neceary to take international migration into account. Thi include the cae in which the creation of job will retain worker who would otherwie have gone abroad or, alternatively, the cae in which foreign killed worker are brought into the country to perform certain ervice. Fourth, the etimation of the EOCL for a project mut conider whether permanent or temporary employment will be created. Temporary poition in ector uch a tourim and contruction lead to greater turnover in the labour market and create condition for voluntary unemployment. Thi churning effect in the labour market reult in additional cot to the economy, which the EOCL hould take into account. Fifth, the rigiditie impoed on the labour market through minimum wage law, retrictive labour practice, and high-wage policie of tate and multinational enterprie in ome countrie tend to create protected ector in the labour market. In uch ituation, quai-voluntary unemployment and eaonal unemployment are common. Under thee circumtance the evaluation of the EOCL ued by a project hould reflect thee pecial labour market condition. Thee five claification within a labour market provide a framework for analyzing the complex concept of EOCL. In thi chapter, the EOCL will be analyzed for the implet cae, i.e., unkilled rural labour. Additional element will then be brought into the analyi a they are needed in order to etimate the EOCL for progreively more complex cae encountered in the appraial of actual project. 342

9 12.4 Economic Opportunity Cot of Unkilled Rural Labour Some well-known growth model of underdeveloped countrie have often taken the mot extreme interpretation of the marginal product forgone hypothei by placing a value of zero on the EOCL of unkilled labour in rural area (Todaro, 1989). A previouly explained, thoe theorie rely upon the aertion that becaue of the large number of unkilled rural worker, there i no economic opportunity cot to filling additional job (Marglin, 1979, pp ). However, there i a lack of empirical evidence for the exitence of a urplu of idle rural labour. In fact, reearch into rural economie provide a peruaive body of evidence indicating that when unkilled worker are not employed in the formal agricultural ector, they pend a large proportion of their time on other productive houehold and family-farming activitie. In thi circumtance, the prevailing daily or weekly wage rate (the upply price of unkilled labour) i a reflection of the marginal productivity of thi type of activity. Hence, the market wage can be ued a an effective meaure of the value of the forgone marginal product of unkilled labour (Harberger, 1971). A erie of tep erve a a guide to the etimation proce when the upply price of labour approach i ued to calculate the EOCL. The firt tep i to determine the minimum gro-of-tax wage (W) needed to attract ufficient unkilled labour to the poition available on the project. The econd i to identify ditortion in the labour market, uch a income taxe or unemployment inurance benefit. The final tep i to determine the EOCL by adjuting the market wage to compenate for uch ditortion. To demontrate thi proce, two cae will be conidered. In the firt cae there are no eaonal variation in either the market wage rate or the demand for unkilled worker. The econd example demontrate the way in which EOCL can be etimated when there are eaonal variation in the market wage rate and in the project demand for unkilled labour over the year. In the firt cae it i aumed that there are no ditortion in the unkilled labour market, i.e., there are no taxe paid by the employer 343

10 (demand ide) and no income taxe paid by the worker (upply ide). It i alo aumed that there are no fluctuation in wage or labour demand over time. It follow that the upply price of labour ( W ) i alway equal to the prevailing market wage (W). Since there are no ditortion, there i no need to make further adjutment to the market wage to etimate the EOCL. Conequently, the market wage rate for unkilled labour i the upply price of labour, which in turn i the EOCL, a hown in equation (12.2). EOCL = W = W (12.2) Note that the EOCL i etimated uing the market upply price ( W ) rather than the project wage ( W p). The project wage i the demand price and meaure the financial cot of labour for a particular project, while the market wage meaure the opportunity cot to the economy of the unkilled labour. If the demand price i higher than the market wage, the difference i an economic externality that arie from the employment of thi type of labour. In the econd cae, the etimation of the EOCL of unkilled labour i carried out for a project that demand worker throughout the year, while the market wage varie a a reult of demand and upply factor affecting the local labour market. Uing the upply price approach, we begin again with the market wage of unkilled labour for thi type of project. There are no tax ditortion. Owing to the eaonal fluctuation in the market wage, the EOCL at any point in time will be calculated by the market wage rate ( W t ) that correpond to the period of time in which labour i hired by the project. For example, if a region growing rice and ugar cane ha a wage rate of $5 per day during the off-eaon, it i poible that the wage could be many time higher during the harveting eaon if thee coincide. If a project i built baed on the aumption that labour will be teadily available at $5 per day, but intead it mut compete for labour at a much higher rate during the harvet eaon, the financial and economic viability of the project may be threatened. Thee higher eaonal labour cot mut be accounted for in arriving 344

11 at an accurate etimate of the EOCL for a project. Furthermore, eaonal variation in the ize of the employed workforce hould alo be reflected in the calculation of the wage. It i a common condition in rural area that both the demand for unkilled labour and the market wage rate have pronounced eaonal pattern, a illutrated in Figure Equation (12.3) deal with thi ituation by defining the total EOCL ued by a project over a year a the product of the quantity of labour hired in each eaon or wage period multiplied by the correponding market wage rate (upply price) for the period. Thi i equal to the um of the unkilled wage rate for each particular eaon or wage period ( W t ) multiplied by the total amount of unkilled labour employed by the project in that period ( L t ) : n ( L t W t ) EOCL = t= 1 (12.3) where t denote the period of time and n denote the total number of period. Figure 12.1: Effect on the of Seaonal Variation in Wage and Labour Demand in Rural Region 345

12 where:.. attern of project demand for labour during the year attern of wage rate for unkilled labour during the year If the project demand for labour i relatively high in the off-eaon, the total EOCL will be lower than if the project demand for labour coincide with the eaonal peak demand for thi labour. Conider the cae of a labour-intenive ugar project. 3 The project require unkilled worker on a temporary bai and pay a wage of $180 ( W ) per month p. The working condition are identical to thoe prevailing in the labour market. Table 12.1 how the project monthly requirement for worker in column (3), and the monthly market wage rate that labour would be willing to work for on thi project in column (2). Table 12.1: Market Wage and roject Demand for Seaonal Labour Month Market Wage ($/month) eron-month Required by roject EOCL for eriod ($) (1) (2) (3) (4) January ,160 February ,800 March ,240 April ,620 May June Example are baed upon work done by Jenkin and El-Hifnawi (1993). 346

13 July Augut September October November ,350 December ,620 Total ,690 In thi cae, the monthly market wage rate are the upply price of unkilled labour to the ugar project. Uing equation (12.3), the EOCL i then calculated a follow: n 12 EOCL = = ( L t W t) t= 1 = = $12,690 The project wage ( W p) paid doe not play a direct role in the etimation of the EOCL. The wage paid by the project i the financial cot to the project. The difference between the financial cot and the EOCL i the value of the labour externality Economic Opportunity Cot of Skilled Labour Skilled labour i not a homogeneou factor, nor are the financial cot and the EOCL going to be the ame for all type of uch labour. There i no doubt that ecuring adequate upplie of labour with the appropriate kill i a key determinant of the ucce of mot project. otevaluation of development invetment have demontrated that project are often eriouly delayed or even abandoned becaue of an inadequate upply of labour with pecific kill. Hence, pecial attention need to be 347

14 paid to determining the ource of upply, level of compenation, and potential ditortion in thee labour market. To meet a project requirement, labour i often induced (with higher wage and a better living environment) to migrate from other area. For example, killed worker in urban area are able to obtain many good and ervice, uch a better education for their children, that are more readily available in the city. If called upon to move from an urban to a rural area, they may well require a wage premium to be paid, in pite of the fact that imple food item are cheaper in the countryide. The upply price approach for determining the EOCL for killed occupation follow the ame baic procedure a the unkilled cae. The firt tep i to determine the market upply price of labour ( W ) needed to attract worker to the project. Ditortion to that wage are then identified and quantified. The EOCL can be etimated by adjuting W to account for thoe ditortion. To demontrate thi approach, the EOCL i etimated for three ituation. The firt example i implified by uing the omewhat unrealitic aumption that there are no ditortion in the market for labour and that the project provide job with the ame working condition a other employer of thee occupation in the area. Furthermore, no worker need to (or can) be attracted from outide the area. The econd cae drop thee aumption and conider a ituation in which labour mut be induced to move from alternative project or region where there are market ditortion. The final cae i one that demontrate how employment that lat for le than a full year can be a factor in determining the value of the EOCL of any particular type of killed labour Labour Market Without Ditortion or Regional Migration If there are no ditortion in the market, uch a income tax on the wage for a given occupation, and if the employment provided by the project ha the ame working condition a alternative employment in the region, it doe not matter whether the new worker come from other employment (reduced demand) or from non-market activitie (new 348

15 upply). In both cae, the EOCL i equivalent to the local market wage (W), which i the upply price ( W ). Thi i exactly the ame reult a in the cae of the unkilled rural labour. In fact, the analyi of the EOCL i differentiated not o much by the kill level of the worker a by the nature of the ditortion in the labour market. In the cae of killed occupation, it i more realitic to aume that a higher wage will have to be paid to attract uch labour away from other job that have different working condition and/or are located in other region that have ditorted labour market Worker Migrate to roject from Ditorted Regional Labour Market Suppoe a project hire labour and ome of the worker are induced to migrate from alternative employment in other labour market. For each occupational type, the project pay a wage equal to or higher than the W gro-of-tax upply price ( g ) to attract an adequate number of worker. A demontrated by Figure 12.2, the migration of worker from the other region to the project will hift the labour upply curve leftward to the new poition S'S' from SS. Thi hift interect the demand curve (DD) at the higher equilibrium wage rate at B from A, cauing a decreae in the demand for the current employment from Q 0 to Q 2. At the ame time, the higher wage rate may induce ome killed worker to enter the formal labour force, or may reult in more overtime being worked, thereby increaing the quantity of killed labour upplied from Q 1 to Q 2. The net effect i that even if all of the labour for the project migrate from the ending region, a proportion of the labour ( ) H ultimately come from the newly induced upply, and a proportion ( H d ) come from the reduced demand for worker elewhere. 4 4 For a further development of ome of thee iue, ee Bell (1991) and Gemmel and app (1991). 349

16 Figure 12.2: Regional Interaction among Skilled Labour Market Quantity of Labour Required for roject Market for a Type of Skilled Labour in Sending Region QQ 2 1 QQ 0 2 H = H d = QQ QQ The reduction in the quantity of labour employed elewhere (i.e., Q 0 Q 2 ) reult in a lo of peronal income taxe to the government, and thi i hown a the area bounded by ABCE; thi i alo the ame a the area meaured by the vertical difference between the gro-of-tax upply curve, SS, and the net-of-tax upply curve, S nsn, multiplied by the change in employment ( Q 0 Q 2). When calculating the EOCL, only the tax lo reulting from the reduced demand ( H d ) need be accounted for becaue it i aumed that the increaed upply ( H ) of labour i 350

17 coming from market or non-market activitie where there are no taxe or other ditortion. Thu, the EOCL for the project in uch cae i the W gro-of-tax upply price ( g ) of worker induced to move to the area minu the difference between the income taxe the worker would pay W T on thi gro-of-tax upply price of labour ( g ), which are gained by the government, and the income taxe previouly paid by the worker in their alternative employment ( H d W a T), which are lot by the government. For implicity ake, it i aumed that the tax rate thee worker pay on the upply price and alternative wage in the ending region are the ame, although thi i not necearily the cae. The EOCL of killed labour hired by the project in the area can be expreed a follow: = (12.4) EOCL WTW g ( g da TWH) where: H d denote the proportion of the project demand for labour obtained from taxed employment activitie in the alternative labour market; W a denote the gro-of-tax wage of labour from alternative ource; W g denote the gro-of-tax upply price of labour; and T denote the income tax rate levied on worker in all region. In thi ituation, H = (1 H d ) include both the upply of labour coming to the region from untaxed market and non-market activitie and increae in labour force participation and the number of hour worked. While it i theoretically poible for a project to change the level of labour force participation or the number of hour worked, over the lifetime of a project thi effect i likely to be mall, depending on the type of kill and the market at the time of recruitment to the project. Let u conider again the ugar project dicued above. In addition to the unkilled worker hired for the project, the project require 1,

18 peron-month of labour in killed occupation each year. Owing to a hortage of uch worker in thi region, the project will have to attract them from the urban area urrounding the region in which the project i located. It i aumed that depite their monthly gro-of-tax alarie ( W a ) of $900 in the urban area, thee worker will not work for the W project for le than $1,200 gro of tax ( g ). Thee wage rate reflect the gro-of-tax upply price of the worker in the two market, repectively. Suppoe there i a policy of encouraging more worker in thee occupation to migrate to the rural area; in thi cae the project i required to pay a higher alary ( W p) of $1,500 per month for uch labour, or $300 more than the upply price. All killed worker pay 20 percent of their wage in income taxe. Uing equation (12.4), we can etimate the EOCL of thi labour to the project by determining: a) the taxe to be paid on the upply price of killed labour for the project and b) the taxe forgone by the worker in their alternative employment. a) Taxe on the Supply rice of Labour W T g = 1,200 (0.20) = $240/month b) Taxe Forgone in Alternative Employment Let u aume that the upply of labour in thee occupation in the economy i relatively inelatic compared with the demand for that labour, and let H d = 0.90 and H = Hence, we can anticipate that approximately 90 percent of the project labour requirement will ultimately be ourced from the decreae in the quantity of labour employed elewhere, while the remaining 10 percent of the project need will be met through increaed labour force participation owing to the new project higher wage. The forgone taxe from the previou employment of the worker are calculated a follow: H d WT a = = $162/month 352

19 Combining thoe two part with the upply price, the EOCL of the labour ued by the project in thi rural area i calculated from equation (12.4) a follow: EOCL = WTW g ( g da TWH) = 1,200 [(1, ) ( )] = $1,122/month The difference between the EOCL and the project wage repreent the value of the project labour externality per month of labour employed. Following equation (12.1), the labour externality for the above cae can be expreed a: LE i= pg + WW( g da TWHTW ) = 1( )1( ) + TWHTWTWTW p g pda Carrying thi analyi one tep further, we can determine how thee labour externalitie are ditributed between the worker and the government. The benefit to each can be calculated a follow: W1( TW ) 1( T) Labour benefit = p g = 1,500 (1 0.20) 1,200 (1 0.20) = $240/month Government benefit = p TW H da TW = 1,500 (0.20) ( ) = $138/month Thu, of the total of externalitie created per month by the employment of labour by a project, the worker will gain an additional $240 per month, while the government will capture $138 per month in additional taxe. The ditributional analyi provide a mean of evaluating the financial gain and loe affecting group in the economy other than the owner of the project. 353

20 12.6 When Labour i Not Employed Full-Time In thi analyi, worker are not divided between thoe who are working in the formal labour market and thoe who are not. Intead, it i potulated that each worker could pend part of each year in non-market activitie or unemployment. In thi cae, worker can expect to be employed in market activitie for a proportion ( p ) of the year if they work for the project. If they are not aociated with the project, they will be employed for a different proportion ( a ) of the year. When they are not working in the formal labour market, they will be engaged in nonmarket activitie outide the project or in alternative region, i.e., (1 p ) and (1 a ) proportion of their labour force time, repectively. The gro-of-tax upply price of killed labour in the area of the W project i again denoted a g and the alternative wage, which reflect thi labour other opportunitie, a W a. From the upply price approach, the EOCL i equal to the gro-of-tax expected upply price for labour W ( g ), but only working a portion of the year on the project ( p ), minu the additional tax payment that the worker would incur if earning their W upply price wage g on thi project. Thi additional tax i the difference between the tax paid on the ( W T) project p g and the tax previouly paid in the alternative mix of market activitie ( H d a W a T). The taxe lot in alternative market activitie arie becaue there i a net reduction in employment of thi type of worker elewhere. We aume that worker do not pay taxe on non-market activitie. Uing the upply price approach, the EOCL of thee worker i the expected gro-of-tax upply price le the expected net change in tax payment. It can be expreed a equation (12.5): 354

21 = (12.5) EOCL pg ( pg ad a TWHTWW ) Suppoe in thi cae that the alternative wage rate for killed labour i W a = $600/month and the project wage i equal to the gro-of-tax upply price paid to induce labour to move to the project area W ( g = Wp = $800/month). The tax rate on killed labour in all location i 20 percent. All of the labour i obtained from alternative employment H = 1), and the proportion of time employed in the alternative area ( d i a = 0.8. Auming that a killed worker expect to be employed in the project and the project region i p = 0.9, the EOCL on thi rural project would be: EOCL = 0.9 (800) [0.9 (800) (0.20) 1.0 (0.8) (600) (0.20)] = 720 (144 96) = $672/month While the financial cot of labour to fill a job (which employ omeone for 90 percent of the year) i etimated on average to be $720 pw (= p ) per month, the EOCL i only $672 per month, or $48 le than the financial cot. Thi difference i the net tax gain to the government. We now extend the analyi to examine a cenario in which worker are employed le than full-time in market activitie during a typical year. Thi i epecially important in the cae of countrie with high unemployment compenation payment, uch a Canada and the countrie of northern Europe. 5 We differentiate between thoe engaged in full-time employment and thoe who have a work hitory 5 In thee countrie, the unemployment benefit vary from 55 percent and 75 percent of lot wage in Canada and Sweden, repectively, to a high a 90 percent of the previou daily wage in Finland. 355

22 characterized by a ucceion of work experience interpered with unemployment. Becaue of their choice of occupation or their level of eniority, people in the permanent (or full-time) employment ector are almot never unemployed. On the other hand, worker employed in temporary ector uch a tourim and contruction are in job that are not expected to be aociated with continuou employment. For thi analyi, individual who are expected to experience periodic pell of unemployment or non-market time are included in the temporary labour force, both when they are working and when they are unemployed. When evaluating project, one further iue for conideration i the quality of the job being created (Jenkin and Kuo, 1978). Job need to be claified by the type of employment they provide. Are the job fulltime for the entire year (i.e., permanent ector), or will they employ a given worker for only part of the year (i.e., temporary job)? Temporary job are thoe that do not retain the worker for a full year but interpere employment with pell of unemployment or non-market activitie. ermanent job provide full-time employment all year round. Identifying the type of employment being created i important becaue temporary job can have a high economic cot when unemployment inurance payment (or other form of ocial ecurity) are paid to uch worker when they are engaged in non-market activitie, including being unemployed (Boadway and Flatter, 1981). Hence, account need to be taken of unemployment inurance in the appraial of a project that create thee job. Let u conider firt the creation of permanent job. When a project create new permanent job, they will generally be filled by individual already working in alternative permanent ector job or other temporary ector job; ome individual will be hired who are currently out of the p t labour force. Thee proportion are denoted a H d, H d, and H, p t repectively, where H d + H d + H = 1. For thoe being ourced from alternative job in the permanent ector, there will be an externality ariing from the lo in income tax receipt from the reduction of employment in thee activitie. For thoe ourced from the temporary ector, there will be a aving in the unemployment inurance being paid to the temporary ector worker when they are unemployed. At the ame time, there i a lo of any taxe they would have paid while working. 356

23 For thoe job that are filled by people who were previouly out of the labour force, no externalitie need to be included. Therefore, the EOCL of a permanent job can be expreed a follow: where: p p EOCL = WT g1( + ) d p TWH t + [ )1( ]) d tt (12.6) TWH 1( t fut W g denote the gro-of-tax upply price of labour to the project; W p denote the gro-of-tax wage earned in alternative job in the permanent ector; W t denote the gro-of-tax wage earned in the temporary ector; t denote the proportion of time a temporary ector worker expect to be employed during a calendar year; T denote the peronal income tax rate; f denote the proportion of time an unemployed worker expect to collect unemployment benefit; and U denote the unemployment inurance benefit. In countrie where there i no unemployment inurance, uch a Indoneia and Vietnam, then f = 0. Equation (12.6a) will then meaure the EOCL to fill a permanent job a: p p t 1( (12.6a) EOCL =) g + dpd + t TWHTWHTW On the other hand, when a year worth of additional employment i created in the temporary ector of a labour market, thee worker will be ourced from the permanent ector, from the temporary ector, and from p t thoe previouly out of the labour force, in the proportion H d, H d, and H, repectively. In thi ituation, uppoe t i the proportion of time that any one peron actually work in a temporary ector job during 357

24 a year. A temporary job are created, and people are attracted to them from the permanent ector, thee people will experience period of unemployment and collect unemployment inurance. Each period of labour ervice ourced from the permanent ector will have aociated with it 1/ t individual and 1( / t) t period of unemployment. Thi / ) 1( t t period of paid unemployment inurance will give rie to compenation. For labour ervice ourced from thoe already in the temporary ector, the lo in taxe will be for the ame length of time a the time working on the project, and the amount of unemployed time and unemployment inurance compenation will alo be the ame a before. The EOCL that relate to a year worth of temporary ector job will then be equal to: T EOCL = + ()1( )/[ TWHTW + fu T])1()1( t g dt pp t + dt TWH [ 1(/) tt ] 1( ) )( + HfU T (12.7) In the cae where the wage rate paid for temporary and permanent job are the ame, the economic cot of 12 month of temporary job would be greater than for a year in a permanent job becaue of the higher amount of taxe that would be lot and the higher amount of unemployment inurance payment aociated with thee job International Migration and the Economic Opportunity Cot of Labour Until recently, labour ha been conidered a non-internationally traded ervice. However, thi i changing a more and more worker are relocating to other countrie to ell their kill and ervice. There are two cenario: retention, or returned migrant, and foreign worker Retained or Returned Migrant Thi i particularly relevant for countrie uch a the hilippine, Egypt, 358

25 and Sri Lanka, where large number of killed and emi-killed worker are regularly employed abroad for ubtantial period of time. In uch a ituation, when a project i created inide the country and additional labour in certain occupation i hired, we would expect to find a part of thi labour to be ourced from a reduction in the outflow of international migration. When thi occur, the EOCL mut take into conideration not only the adjutment of the demand and upply of labour in the local market, but alo any ditortion aociated with the retention or return of migrant who would have been employed abroad. It i quite common for a country citizen who are working abroad to end back a tream of payment in the form of peronal aving or remittance to relative. Following the upply price approach to the EOCL, reduction in remittance are not an economic cot a they will be factored into the worker upply price to the project. However, an adjutment need to be made to the upply price becaue the remittance are made in foreign exchange, in which a foreign exchange premium exit in mot countrie. Taking into account the adjutment to both the local and the international labour market, the expreion for the EOCL become: EOCL = W g (1 T) + H d W a T + H f R( E e E m 1) (12.8) where: H d denote the proportion of the project demand for a given type of labour obtained from taxed employment activitie in the dometic market; H f denote the proportion of the project demand for a given type of labour ourced from reduced international out-migration; R denote the average amount of remittance (meaured in local currency) that would have been made per period if thi type of worker had been employed abroad; and e E ( 1) m E denote the rate of foreign exchange premium a a fraction of the value of the amount that would otherwie have been remitted. 359

26 When ome of the ourcing of labour for a project i through an adjutment in the international migration of worker, it i recognized that a hare of thi labour i being ourced from alternative dometic activitie ( H d = 0.6) and a hare from change in international flow of labour H f = 0.3. Let u alo aume that the VAT rate i 15 percent, that thee worker would have remitted $500 on average per month, and the foreign exchange premium i 6 percent. Applying equation (12.8), the EOCL i a follow: EOCL = W g (1 T) + H d W a T + H f R( E e E 1) m = 1,200 (1 0.2) (900) (500) (0.06) = $1,077 The difference between the EOCL and the project wage repreent the value of the project labour externality including the premium of foreign exchange that i no longer remitted to the project country Foreign Labour In countrie where the labour hortage i particularly acute, it may be neceary to import foreign worker to work on project. Example of thi practice can be een in both developing and developed countrie where the demand for labour exceed the upply. Often, foreign worker are brought into the country by corporation or government to work on project requiring their kill. In developing countrie, thi often take the form of killed advier or technical taff, while in developed countrie, guet worker or unkilled labourer are imported to fill gap in the labour pool. There i an EOCL aociated with thi foreign labour f ( EOCL ), which hould be included in the project aement. f EOCL i the net-of-tax wage paid to the foreign worker plu adjutment to the amount of foreign exchange aociated with the repatriated portion, and adjutment to the amount of VAT aociated 360

27 with conumption by the foreign worker uing the non-repatriated portion of that wage, plu any ubidie the foreign worker may benefit from while in the country. The repatriated portion of the wage hould be adjuted to account for the true cot of the foreign exchange to the economy rather than jut it market value. Thi i neceary becaue the value of the foreign exchange may be ditorted. While living in the country, foreign worker have to ue a portion of their wage for conumption. The incremental amount of VAT revenue paid a a reult of the foreign worker conumption in the country hould be accounted for a an economic benefit to the country, a the country gain from the local conumption of the foreign worker. At the ame time, foreign worker may benefit from government ubidie on a variety of item, uch a food, fuel, houing, and health care. The amount of benefit received by foreign labour from thoe ubidie hould be accounted for a an economic cot to the country. The EOCL of foreign labour can be expreed a: f f f 1( h)1( TWTW h )1)( tr VAT e f E 1( TR h)( m + N EOCL = + W )1 E f (12.9) where: W denote the gro-of-tax wage of foreign labour; T h denote the peronal income tax levied by the hot country on foreign labour; t VAT denote the VAT rate levied on conumption; R denote the proportion of the net-of-tax income repatriated by foreign labour; e E denote the economic cot of foreign exchange; m E denote the market exchange rate; and N denote the value of benefit gained by foreign worker from ubidie. f If EOCL i greater than the financial cot of labour to the project, the econd term mut be maller than the um of the third and fourth 361

28 term; thi implie that the economic benefit created by foreign conumption in the country cannot offet the foreign exchange premium related to the remitted portion of the wage and the cot of government ubidie. In thi cae, the EOCL of hiring foreign labour will be greater than the project wage. However, if the econd term i greater than the third and fourth term, the EOCL of foreign labour will actually be lower than the market wage; thi mean that the country i benefiting economically from the preence of foreign labour. Suppoe a multinational corporation conider an electronicaembly project in an urban area and dicover that there i inufficient local labour. It decide to import killed worker from a nearby country to operate the project until enough local worker can be trained for the production requirement. The hortfall i etimated to be equivalent to 50 worker, who will be paid $2,000 per month. That wage will be ubject to a 25 percent peronal income tax. Worker are expected to repatriate 30 percent of their net-of-tax income to upport family member at home. The VAT rate i 15 percent. The market exchange rate i held contant by the government, while the economic exchange rate i etimated to be 6 percent higher than the market value. In thi cae, it i aumed that no ubidie are paid by the government with repect to thee worker, i.e., N = 0. Applying thee value to equation (12.9), the EOCL of foreign labour i: f EOCL = 2,000 (1 0.25) 2,000 (1 0.25) (1 0.30) (0.15) + 2,000 (1 0.25) (0.30) (0.06) = $1,369.50/month Thi analyi how that the EOCL of each worker will be $ le than the gro-of-tax wage of $2,000. Hence, a ubtantial external benefit i generated by thi ue of foreign labour Effect of a rotected Sector on the Economic Opportunity Cot of Labour 362

29 The focu o far ha been on etimating the EOCL in competitive labour market. In many countrie, however, the urban labour market i egmented into a protected ector and an unprotected, or open, ector. 6 The protected ector i uually made up of the government agencie, foreign companie, and large local firm that provide wage ( W ) above the market-clearing wage. The higher wage offered by thee type of employer are often the reult of tricter compliance with minimum wage law, powerful union that are able to demand and receive ignificantly higher wage, government policie that give higher wage to civil ervant, or foreign companie that pay high wage to decreae poible reentment by worker and politician in the hot country. Conequently, employment in the protected urban labour force i highly attractive, with a variety of rationing method being ued to elect the people to fill the limited number of poition. The open labour market i typically affected by fewer ditortion to O the upply price of labour ( W ). Wage are determined competitively in the marketplace, where there are fewer barrier to entry, lower wage, and le ecurity of employment. While worker may be initially attracted to thi labour market by the propect of finding a job in the protected ector, they often end up working in the open labour market. The phenomenon of chronic unemployment, at rate far in exce of what might be explained in term of normal friction in the economy, ha been attributed, in part, to the exitence of a protected labour market. A portion of thoe chronically unemployed worker are attempting to gain acce to the protected ector but, at the ame time, are unwilling to work for the lower wage offered by the open labour market. Thi unwillingne to work for the open market wage create ub-ector in the labour market in which quai-voluntary and earch unemployment exit. 6 The dicuion of the EOCL for the protected ector tart from the approach followed by Edward (1989). See alo Bicak et al. (2004). 363

30 The rotected Sector and No Migration The characteritic of unemployment in thi ituation are hown in Figure 12.3A. If the overall upply of labour to the market i given by T the upply curve ( SS ), the total number of worker making themelve available for work at the protected ector wage of W 1 i hown a point C. The number of protected ector job available i much more limited at Q (i.e., BC). Hence, there i an exce upply of labour available at the protected ector wage, a hown by the quantity B. If the election of worker for employment in the protected ector i carried out in a random fahion from the available worker, independent of their upply price, it follow that the upply of labour available to the open market T will be a fraction (B/C) of the total labour upply SS at each wage rate. O Thi labour upply i hown a the curve SS. To implify the analyi for thi cae, it i aumed that the demand O for labour in the open ector i perfectly elatic at a wage rate of W and that the interection of the demand for labour in the open ector ( O D O O W ) with the upply ( SS ) determine the quantity employed in the open market. Thi quantity i indicated by point A 1. The quantity of labour claified a unemployed ( Q QV ) i determined by the difference between point A 1 and B. Thee quai-voluntary unemployed are thoe worker who will chooe not to take job in the open market ector becaue their baic upply price of labour i above the open market O wage ( W ). They actively eek job in the protected ector and will conider themelve involuntarily unemployed. They are eeking work that will pay the protected ector wage ( W 1 ), but are unable to find it. If a project i added to the protected ector, then, a hown in Figure 12.3B, the ize of the protected ector increae from (C B) to (C B 1 ). If, again, thee additional worker (B B 1 ) are elected randomly from thoe remaining who want to work in the protected ector, O the upply of labour to the open market will hift to the left, from SS 364