Many developing and emerging economies have undertaken over the past decades major reforms on their product and labor markets. The impact of these ...
Preliminary draft, please do not quote
Product and labor market reforms in an economy with rents, bargaining and informality May 2005 Lodovico Pizzati and Stefano Scarpetta
ABSTRACT
Many developing and emerging economies have undertaken over the past decades major reforms on their product and labor markets. The impact of these reforms on size and distribution of market rents has been influenced by the presence of a significant informal sector. This paper presents a simple general equilibrium model that illustrates the possible links between product and labor market reforms, informality and labor market outcomes. The main features of the model are: (a) a two-sector economy with a formal and informal economy; (b) imperfect competition in the product market affecting firms’ profit margins; (c) an endogenous informal sector with firms entry and exit; (d) a twofactor production function with skilled and unskilled workers; (e) labor market regulations such as a minimum wage, union bargaining, and an employment tax; and (f) firm-specific technologies affecting each firm’s share of skilled and unskilled workers. With this simple framework we simulate the effects of product and labor market reforms using data from eight countries with different economic conditions and policy settings: Cameroon, Chile, Colombia, Honduras, Indonesia, Paraguay, Poland, Russia. The simulations suggest that when firms can choose between the formal and the informal sector the short-term impact of reforms may differ significantly from that in the long-run. Also, initial conditions such as size of informal economy, the share of unskilled workers, and wage differentials can lead to different effects of similar reforms across countries.
1
1.
Introduction Does the presence of a large informal economy affect the way in which countries
adjust to product and labor market reforms? This question is of importance given the fact that many developing and emerging economies characterized by a sizeable informal economy have embarked over the past decades on ambitious process of reforms – ranging from macro stabilization measures, to liberalization of trade and financial markets, to privatization and de-regulations of product and, seldom, labor market. While many of them have experienced some improvements in long-term sustainable growth, labor market outcomes have often been disappointing with the persistence of high unemployment or under-employment. Moreover, the informal sector has often increased from already high levels, probably absorbing firms and workers whose performances were no longer viable in the more competitive but also more regulated and costly formal sector. Productivity-enhancing reforms are expected to generate changes in market rents and a redistribution of these rents among economic actors, who will react to these changes by modifying their behavior (see Blanchard and Giavazzi, 2002 for a discussion on this point). The type of reforms is also likely to influence both the size of the rent change as well the magnitude of the redistribution of these rents. Product market reforms are likely to affect predominantly the size of the rent while reforms of the labor market will affect predominantly the distribution of rents between firms and workers.
In
developing countries, the option of moving to the informal market also gives to both workers and firms another path of adjustment to reforms in the formal sector. The easier for firms to stay “below the radar screen” the closer the possible interactions between formal and informal activities and thus the higher the potential shifts of firms between them after a reform. Understanding the dynamic process that reforms put in motion and its short-term and long-term effects on firms and workers is of great importance to assess its potential benefits and costs. For example, product market reforms may force some firms to exit the market – generating unemployment – or shifts of firms and workers to the un-regulated economy. Labor reforms may reduce bargaining power of workers and shift rents to firms without much effect on formal employment. But joint product and labor reforms may promote formalization of both firms and employment. These different 2
dynamic adjustments to reforms have profound implications for the level and type of employment, wages for different workers and ultimately poverty. In this paper we shed light on the dynamic adjustments associated with different types of reforms using an illustrative general equilibrium model.
The model is
sufficiently simple to clearly show the sequence of changes taking place in the economy, but should be considered as an illustrative tool rather than a precise instrument for policy simulation. Following Blanchard and Giavazzi (2002), we assume an economy with regulations in both product and labor markets. Firms interact under an imperfect competition framework. We depart from Blanchard and Giavazzi (2002) by assuming that the economy features a formal and an informal sector. In the short run firms cannot move from one sector to the other, and only labor can be reallocated. Over the long-run, the number of firms operating in the informal sector is determined endogenously. A second significant feature of this model is that firms are not identical but possess different technology levels. This idiosyncrasy in productivity is reflected by the degree of skillintensiveness employed by each firm that is proxied by the composition of employment between skilled and un-skilled workers, which in turn depends on firm-specific technologies. This two-factor specification is crucial for capturing the wage gap between skilled and unskilled, how it relates to the size of the informal sector, and the direct and indirect effects of policy reforms. The paper is organized as follows. The next section discusses briefly how the presence of a large and persistent informal sector affects the design of a macro model of the economy and the labor market. The following four sections present our model, progressing in steps. Section 3 introduces the baseline model with imperfect competition in the product market, with the degree of competition determining firms’ profit margins. Each firm has a different technology related to its skill-intensiveness, which determines, along with formal economy regulations, firms’ profitability. Consequently, in a twosector model, each firm chooses to be in either the formal or informal sector. Section 4 extends the baseline model by introducing a two-factor production, with firm-specific shares of skilled and unskilled workers. Unskilled workers in formal economy are paid at the exogenously-defined minimum wage; those not employed in the formal sector at that wage are employed in the informal sector at a market clearing wage; while wages for skilled workers are determined by a bargaining process between unions and employers, 3
leading to unemployment among skilled workers. Section 5 discusses two distinct model closures with tax revenues financing either government spending or unemployment benefits. Once the analytical framework is specified, Section 6 presents four illustrative policy scenarios: a product market reforms that promote competition and reduce market rents; an increase in the tax of labor use; an increase in the minimum wage; and a reduction in union bargaining power. In the short run labor movements, in and out of the informal economy, are influenced by what happens to the wage gap between skilled and unskilled workers. In the long run the same policy causes firms’ migration in or out of informality, often countering the short run impact. Clearly the impact of these policies will depend on initial conditions such as the size of the informal economy, the share of unskilled workers, and the wage gap between skilled and unskilled. To she light on Therefore, we parametrize the model using real data for eight different economies (Cameroon, Chile, Colombia, Honduras, Indonesia, Paraguay, Poland, Russia) and discuss the different effects across countries of similar policy reforms. Section 7 concludes.
4
2.
Product and labor reforms when informality looms large Many developing and emerging economies have embarked over the past decades
in far-reaching reforms – from macro-stabilization measures, to liberalization of trade and financial market, to the privatization of many firms to regulatory reforms. While the aggregate economic impacts of these reforms have differed across countries depending on initial conditions and on the way such packages were designed and implemented, in many cases their implementation has been accompanied by rises in unemployment, underemployment, and inequality. In the transition economies of Central and Eastern Europe unemployment was practically nil at the beginning of the 1990s and it is still above 10 percent even after a prolonged period of strong economic growth. Many Latin American countries, including those with sustained growth, have also seen major rises in unemployment and under-employment in the informal sector together with falls in participation rates. At the same time, while measured unemployment has remained relatively low in Sub-Saharan Africa and in South Asia, the share of working poor in the informal sector has reached almost 40% on average in both regions. Many factors may contribute to explain these un-satisfactory outcomes, including the difficult initial conditions, the deep and scope of reforms and more generally their ability to promote a better business environment. But a general feature of most reforms is that they involve a transitional dynamics: labor and capital have to be reallocated across firms, sectors and geographical locations, and the reallocation may be painful for both firms and workers involved. Some firms may be better off as they gain access to broader markets and technologies, and skilled workers may have access to better jobs, but other firms may be forced out of business leading to job losses. The presence of a large informal sector may also affect the effectiveness of reforms and the transitional dynamics. This largely depends on the nature of the informal sector. The official definition of informality also varies: the ILO (2002) refers to the informal sector as composed of non-professional self employed, domestic workers, unpaid workers and employers in firms with fewer than 5 employees.
In many
low-income countries, this definition encompasses the large majority of the workforce, with only public-sector employees and the few employed in large firms being recorded as formal workers. In a number of Latin American countries, formality is related to being
5
affiliated to the social security system,1 while in several transition economies of Central and Eastern Europe and the Former Soviet Union, many workers either receive income support as unemployed and do perform some activity in the unrecorded economy, or have two jobs, one in the formal sector for a relatively low pay and one in the informal sector to provide additional income to the household. Traditional models of duality in the labor market in developing countries depict the informal sector as a free entry sector of last resort with little links with the formal sector (see Fields, 2004 for a review duality models). Recent evidence from Latin America challenges this view of dualism in the labor market (see e.g. IADB, 2003; Maloney 2003) by noticing large flows of workers between the formal and the informal sector and strong spillover effects of wages from the formal to the informal sector. The IADB (2003) reported that in any given six months period 16 percent of workers in Mexico and 11 percent of workers in Argentina moved from the formal to the informal sector or vice versa. And Maloney and Nunez (2004) suggest that, for example, the minimum wage in the formal sector acts as a strong pay signal for the informal sector. In this paper, we follow the view that the informal sector has its own internal duality. Some informal activities are very close to the formal ones, and some workers may actually opt out of formality voluntarily because of higher earnings, more freedom and especially because of the perceived un-reliability of social protection mechanisms in the formal sector; but other workers are trapped into low-productive informal jobs (Fields, 1990; Ranis and Stewards, 1999). Similarly some firms may opt out of the formal sector if the costs of formality are too high compared with the expected benefits related to formality that may depend on optimal firm size and easier access to markets, services and even more skilled workers. An important issue in this characterization of the labor market is how the different segments of the labor market interact. Different hypotheses have been formulated in the literature, ranging from the fully integrated market model with market clearing, to multisector models with wage differential but no unemployment, to different versions of the Harris-Todaro model with wage differentials and unemployment. In our model, we assume five labor market states: i) high-skilled employment in the formal sector with 1
See Auerbach et al. (2005) for evidence of affiliation to social security in a number of countries in Latin America.
6
bargaining power; ii) low skilled employment in the formal sector, with a minimum wage; iii) high skilled employed or self-employed in the informal sector (the voluntary informal); iv) low-skilled workers in the informal sector with market clearing wages (the involuntary informal); and v) unemployment in the formal sector. We assume that total skilled employment and skilled wages are determined as a mark-up of the reservation wage, with the size of the mark-up depending on workers’ bargaining power. In our right-to-manage framework, once wages are set, firms choose the optimal level of skilled employment to maximize profits. If the reservation wage is above the market clearing level, some of the skilled workers will end up unemployed. Moreover, more skilled workers search for formal sector jobs than those who can be hired by formal firms, the other will be employed by informal firms. Labor demand in the formal sector for unskilled workers depends on the number of formal firms, their optimal combination of skilled and unskilled labor, and the level of the minimum wage. Those employed in the formal sector will receive the minimum wage; the others will be employed in the informal sector at the market-clearing wage. Our model assumes no unemployment among the low skilled. We believe that this characterization of the labor market suits well many developing and emerging economies with sizeable informal economy. In the following sections, we introduce our illustrative model, starting with its most simple version in which there is only one type of labor, and then introducing two types of labor and regulations in the labor market as well as different uses of the tax receipts.
7
3.
A Two-Sector Economy Following Blanchard and Giavazzi (2003), we characterize our economy as made
of firms producing differentiated products with monopolistic competition in the goods market, which determines a price mark-up over cost. We also consider two periods: the short run in which firm’s location in the formal (informal) sector is given; and a long run, in which firms can choose between the formal and the informal sector. We start with a simplified version of the model with homogeneous labor. There are n firms in the economy, f of which are formal and the rest are informal. Households have identical preferences over the produced n goods. Consumption for each good, Ci , depends on the utility function shown in Equation (1): 1
(1)
MAX Ci
n ρ ρ C = n ρ −1 ∑ C i i =1
n
subject to
∑P ⋅C i =1
i
i
=Y
Given a budget constraint of income, Y, and prices for each good, Pi , we derive the following demand functions: (2)
Ci 1 P = ⋅ C n Pi
σ
where σ =
1 >1 1− ρ
The resulting consumption demand for good i, relative to the aggregate consumption bundle C, is an inverse function of its price Pi, relative to the aggregate price level P. Each demand curve is characterized by the elasticity of substitution between goods, σ.2 While consumers’ preferences for each good are identical, firms have differing technologies. Each firm’s production, Xi is a function of employment, Ni and a technology parameter ki, which differs in each firm. (3)
2
X i = ki ⋅ N i
ρ 1 n By duality, the aggregate price level is: P = ∑ Pi ρ −1 n i =1
8
ρ −1 ρ
where P ⋅ C =
n
∑P ⋅C i =1
i
i
.
The higher ki the lower the level of employment necessary for firm i to attain a given level of production. So, more productive firms will be less labor intensive for a given level of production. In this section, the distinction between formal and informal economy is determined by labor market regulations, in particular a minimum wage that may be above the market clearing level. This hypothesis is consistent with the evidence in many lowand middle-income countries of a minimum wage which is often much higher than the average wage in the economy (formal and informal). These minimum wages often set the basis remuneration for public sector workers rather than imposing a wage floor for the economy.
Those in the informal sector generally receive only a fraction of these
normative wages. Therefore, for the purposes of this section we assume firms in the formal sector pay workers a formal wage, WF, which isexogenously determined by labor market regulations. In addition, the labor market has full-employment (unemployment is introduced in the next section), as the portion of the labor force, L, that is not employed in the formal sector, NF, will find employment in the informal economy, NI. Therefore, firms in the informal economy, pay market-clearing wages, Wi. The next section will discuss in more detail labor market regulations, union bargaining, unemployment, and the role of the reservation wage. (4)
NI = L −NF
where N F = ∑i =1 N iF f
and
N I = ∑i = f +1 N iI n− f
In conclusion, profits for firm i, will depend on which sector the firm belongs to (Equations (5a) and (5b)), and profits and wage revenues are aggregated into the representative household’s income (Equation (6)), which serves as the budget constraint in Equation (1). (5a)
Π iF = Pi ⋅ X i − WF ⋅ N i
(6)
Y = Π F + Π I + WF ⋅ N F + ∑i = f +1Wi ⋅ N iI
(5b)
Π iI = Pi ⋅ X i − Wi ⋅ N i
n− f
where Π F = ∑i Π iF and
9
Π I = ∑i Π iI
Short-Run Partial Equilibrium Firms make employment and pricing decisions as profit maximizing agents. Given a production level, employment is determined by Equation (3), so that Ni = Xi / ki . Therefore, substituting Equation (3) for Ni in Equations (5a) and (5b) makes profits a function of price, Pi, and production, Xi. Now, because of goods market equilibrium, for each good i production equals consumption ( Ci = Xi ). Therefore substituting Equation (2) for Xi makes profits just a function price, Pi. Under monopolistic competition, firms choose the price level that maximizes profits. Whether in the formal or informal sector the resulting price level is a mark-up over per-unit costs: (7a)
Pi =
σ
WF σ − 1 ki ⋅
(7b)
Pi =
σ
Wi σ − 1 ki ⋅
The mark-up is a function of the elasticity of substitution between goods. This implies that the higher the competition in an economy (greater elasticity), the smaller the mark-up over per-unit costs. Short-Run General Equilibrium Under partial equilibrium each firm chose price and employment levels independently
from
one
another.
Under
general
equilibrium,
however,
the
interdependence between economic agents is taken into account. Since consumption of each good enters the utility function symmetrically, the representative household will consume each product in equal amounts ( Ci =C/n for all i’s). Using this condition into Equation (7), the short-run general equilibrium price level is the same for all products (Pi =P for all i’s) and it is normalized to 1.3 Substituting Pi = 1 in Equation (7b) determines wages in informal economy. Per unit-cost in the informal sector (Wi/ki ) becomes: (8)
Wi σ − 1 = =ρ σ ki
It should be noted that firms have differing technologies and they will offer a different informal wage rate, the least productive firms providing the lowest wage rates.
3
The assumption that consumers value products equally may seem unrealistic when modeling products of different technology. However, relaxing this assumption just introduces an heterogeneity in consumption and prices without changing the general equilibrium implications.
10
Long-Run General Equilibrium While in the short-run firms cannot change sector, in the long-run they may choose to move into the informal economy (or vice versa) depending on which sector is more profitable. If ΠiF > ΠiI , then firm i chooses the formal sector. Substituting Equation (3) into (5a) and (5b) we get: W (9a) Π iF > Π iI → X i ⋅ Pi − F ki
W > X i ⋅ Pi − i → WF < Wi ki
The long-run condition shown in Equation (9a) simply states that if the fixed formal wage, WF, is lower than the wage a firm would otherwise pay in the informal economy, Wi, then that firm will choose the formal sector. The distribution of Wi’s depends on the technology assumption, as more productive firms can afford to pay higher wages and still maintain the profit mark-up. Therefore, while for productive firms the fixed formal wage is preferable, for less productive firms it is too high (see Figure 1). Substituting (8) into (9a) we get an alternate long-run condition: WF INFORMAL 11
firms
4.
A Two-Factor Economy
In the previous section, differences in technology were assumed as given. This section relaxes this assumption by linking the level of technology of each firm to the skill-intensiveness of production. In doing so the assumption of differing technology levels among firms is now justified by relating it on the share of skilled workers being employed. Equation (3) is now substituted by a Cobb-Douglas function that takes into account the employment of two types of workers, skilled and unskilled: (10)
α
X i = k (α i ) ⋅ NU ,i i ⋅ N S ,i
1−α i
The parameter α is firm specific, so that each firm employs a different share of skilled and unskilled workers in order to achieve the same level of output. In addition, the productivity parameter k is now a function of α, implying that the more skilled-intensive a firm is, the more productive it is.4 As two set of workers (skilled and unskilled) are now introduced, this section also makes different assumptions about wage setting in the two respective labor markets: i. Skilled Workers The wage rate for skilled workers is determined by union-firms right-to-manage bargaining,5 and is set as a mark-up over a reservation wage, WR: (11)
β ⋅ WR WS = 1 + (σ − 1) ⋅ (1 − α )
The mark-up increases with union’s power, which is captured by the parameter, β, and decreases with the goods market competitiveness, σ. In addition, the mark-up increases the more skill-intensive the representative (or average) formal firm is, captured by the parameter α .6 The reservation wage, WR, is a decreasing function of skilled unemployment, US, capturing the notion that higher unemployment makes it more painful
k(⋅) > 0 , and k’
