ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

An Inquiry into the Theoretical Structure Underlying the Labour Market Flexibility Argument in India

Anamitra Roychowdhury ( teaches at the Centre for Informal Sector and Labour Studies, Jawaharlal Nehru University, New Delhi.

The theory underlying the labour market flexibility argument in India is identified and critically examined. A review of the empirical studies reveals that this argument is based on rigid real wages, in turn, explained by hiring and firing costs. Such explanations are provided by the insider–outsider theory of employment and unemployment due to Lindbeck and Snower (1988). A scrutiny of the I–O theory reveals that it could not even explain the existence of involuntary unemployment under reasonable assumptions. Further, it is shown that its policy recommendation necessarily assumes Say’s law. Thus, it is concluded the theoretical foundation underlying the LMF is unsound.

The author wishes to thank Prabhat Patnaik, Nivedita Sarkar and the participants at the 55th Annual Conference of the Indian Society of Labour Economics, 2013 for their helpful comments.

India’s labour laws have evolved in a manner which has greatly reduced the flexibility available to the employers to adjust the [organised] labour force in the light of changing economic circumstances …Some changes in the laws are therefore necessary if we want to see rapid [economic] growth. (Government of India 2001: 171)

The above statement captures the outlook held by the Indian government about the organised labour market in India. In this view, rigid labour laws (requiring firms to seek prior permission from the government to fireworkers) are identified to be the primary reason for both the minute share of organised sector1 workforce in total workforce (approximately 8%) and the slow growth in organised employment2 (and output). Consequently, labour laws restricting free hire and fire policy are sought to be relaxed by introducing labour market flexibility (LMF). Only a cursory look at the government documents testifies to this fact. For example, the terms of reference of the Second National Commission on Labour (SNCL), 2002 clearly states that one of the two key tasks3 the commission set itself to resolve was, “to suggest
rationalisation of existing laws relating to labour in the organised sector” (SNCL 2002).

However, anybody acquainted with the literature on LMF in India clearly realises the lack of proper theoretical4 understanding of the subject. It is precisely due to this reason that most of the debate in the Indian context has remained confined to the empirical terrain. The theory which suggests the policy conclusion of complete freedom to hire and fire, especially in the Indian context, is not even clearly stated (let alone scrutinised) in the literature. At the same time, many empirical studies are conducted with an assumption that the underlying theory is unambiguously true. The present paper tries to close this gap in the literature.

Identifying the Theory Underlying LMF

Unfortunately, as noted above, the theory leading to the policy recommendation of unlimited freedom to hire and fire has not been clearly stated in the literature. To locate the theory underlying LMF, we carefully follow here the explanations provided by the empirical studies in interpreting their results. Thus, the robustness of the empirical results or the methodology adopted by them to arrive at those results is not of concern in this paper; rather their underlying theoretical framework is the object of inquiry.

Fallon and Lucas (1991, 1993) are the most influential empirical studies arguing for the introduction of LMF in India.5 They argue that job security regulation (JSR) acts as an impediment to employment growth; in particular, “[by] restricting employers’ ability to fire workers [JSR] may actually reduce the size of the workforce employers wish to maintain” (Fallon and Lucas 1991). The authors explain this reduced ability to fireworkers in terms of increased adjustment costs or labour turnover costs (LTCs)6 due to the 1976 amendment to the Industrial Disputes Act, 1947. They argue that labour adjustment costs negatively impact firms’ employment decision in the following manner:

By raising turnover costs, … job security regulations may thus act as a tax on employment, effectively raising the cost of labour. Job security regulations would then tend to encourage adoption of more capital intensive production techniques, to shift production away from more labour intensive processes, and hence to reduce the demand for labour. (emphasis added; Fallon and Lucas 1991)

Now the question arises: how do LTCs or adjustment costs raise the cost of labour? The authors argue that the JSR secures the position of the currently employed workers (known as insiders) and increases their bargaining power, translating to higher wage claims.7 The rise in wages post regulation is explained in terms of, “This [empirical] finding would be consistent with the theory that enhanced job security endows insiders with greater bargaining strength, thus enabling them to achieve higher wages” (emphasis added; Fallon and Lucas 1991).

Thus, they are using a theoretical framework in which JSR introduces LTCs, which, in turn, is functional in enhancing the bargaining power of insiders translating to their higher real wage claims, thereby raising the cost of labour.8 Since the real wage is inversely related to employment, this, therefore, explains the existence of involuntary unemployment.

Another vital contribution to the literature is the study by Besley and Burgess (2004) that looks into the impact of labour laws (state-level amendments) on employment. In their study, the authors point out two main theoretical routes through which labour regulation adversely affects employment creation: (a) relative price effect, and (b) expropriation effect.

With regards to the first, the following impact is pointed out:

The relative price effect is relevant if the effect of labour regulation is to raise the (fixed or marginal) cost of employing labourers. Labour regulations will typically create adjustment costs in hiring and firing labour and in making adjustments in the organisation of production. We would expect firms in the registered sector to substitute away from labour (reducing employment) towards other labour saving inputs (including capital if labour and capital are substitutes) (emphasis added; Besley and Burgess 2004).

Hence, once again we find that the negative impact of labour laws on employment creation is explained in terms of adjustment (or labour turnover) costs.

In explaining the second route, they argued,

The expropriation effect ... by increasing the bargaining power of the workers, … [causes firms to] face a problem if workers can expropriate part of that return [hence pushing up wages] once the capital is sunk. … This effect shows why pro-worker labour regulations are similar to insecure property rights for owners of capital as their sunk investments are subject to workers expropriation. (Besley and Burgess 2004)

Thus, rent-seeking activities by workers are used to explain higher wage claims—over and above their supply price.9 Now the bargaining power of worker enhances (leading to rent-seeking activity) following the institution of JSR—presumably due to the LTCs associated with hiring and firing.

Therefore, as before, we find that labour regulations impair employment creation (forestalling achievement of full employment) by amplifying turnover costs (thus acting against the interest of the currently10 unemployed). The existing workers use this regulation in their favour to claim higher wages for themselves.11 Thus, pro-labour regulation, in the theory operates through introducing LTCs, which secures the position of the current workers against free dismissal. Knowing this fact, the currently employed workers immediately claim higher wages (up to the extent of LTCs)—since the firm has to bear LTCs in case they are replaced. Let RW denote the reservation wage of currently unemployed workers. Let “h” be the marginal hiring cost of labour and “f” be the marginal firing cost of labour. Now, it is argued that the currently employed workers manipulate to increase their wages (WI) above the reservation wage of the currently unemployed workers by the amount of marginal hiring and firing costs (that is, WI= RW + h + f). However, the firm would be indifferent between retrenching an existing worker and employing a hitherto unemployed worker (drawing on the pool of unemployed workers) in their place; for, the cost of replacing an existing worker by a new worker would be, RW + h + f, exactly the same as retaining an existing worker.

Thus, JSR (through LTCs) is instrumental in raising the real wages of the existing workers. Hence, LTCs provide the necessary bargaining tool to the existing workers for increasing their wages. Therefore, the theory explains the source of bargaining power of the existing workers in a firm in terms of labour adjustment/turnover costs.

Such an explanation is provided by the insider–outsider (I–O) theory of employment and unemployment propounded by Assar Lindbeck and Dennis J Snower—which explains how LTC is instrumental in raising the real wages of the existing workers12 (above market clearing levels). It is this explanation of real wage rigidity13 and the prevention of its underbidding by the currently unemployed, which explains involuntary
unemployment. In the authors’ words,

In the simple models presented in this book, we have seen that the insider wage is … positively related to the magnitude of labor turnover costs … Moreover, a rise in the insider wage may be expected to lead to a rise in the level of unemployment, … because labor demand in the primary sector may be inversely related to the insider wage … (Lindbeck and Snower 1988).

Therefore, like all neoclassical explanations of involuntary unemployment, the I–O theory also tries to provide an answer solely based on the labour market and quite predictably explains it through rigid14 real wages set above the market clearing levels15 by the LTC. That this line of reasoning is banked upon by the empirical studies in the Indian context to explain rising real wages, which in turn is held responsible for generating unemployment, is not difficult to establish. Ghose (2005) notes,

The sharp rise in real wage in the 1980s could be conceivably be attributed to the introduction of strict job security regulations in 1982. By making labour retrenchment practically impossible, these regulations suddenly increased the bargaining power of the workers in employment in organised manufacturing and effectively created an “insider-outsider” problem. It is possible to suppose that the workers used this increased bargaining power to increase real wages. (emphasis added)

Similarly, Fallon and Lucas (1991) observes,

The inability of employers to dismiss striking workers may strengthen the bargaining position of unions and insider workers more generally, which could result in an enhanced overall reward package including both greater job security and higher pay. (emphasis added)

Moreover, discussing the various arguments forwarded in the literature to explain employment slowdown in the 1980s, Dutta Roy (2004) noted—“Ahluwalia (1991) and the World Bank (1989), on the other hand, explain the employment decline as the impact of a high rate of growth of wages, in turn made possible by job security legislation.” In the footnote to this sentence, the author comments, “A possible basis for the latter possibility is provided by the Insider–Outsider framework of Lindbeck and Snower (1988)” (Dutta Roy 2004). Further, D’Souza (2010) notes, “The insiders who are privileged to have job security, however, are those who have been able to capture rents [manipulating their wages] as firing costs are high when job security legislation is in place.”

However, the explanation of unemployment provided above—in terms of rent-seeking activity of the currently employed workers (insiders)—is somewhat misleading due to the following reason. In light of LTCs, although it is true that firms are indifferent between paying higher wages to insiders (above the supply price of labour) and hiring an outsider in their place; however, insiders may not be in a position to exploit this fact in their favour, precisely because there is unemployment in the economy. Let us elaborate on this point.

Insiders and Outsiders

It is clear that insiders prefer their current position of being employed rather than joining the ranks of outsiders. Knowing this fact, firms—in order to teach insiders a lesson and contain their higher wage claims—replace some of its insiders by outsiders. Firms can easily do this for they are indifferent between insiders and outsiders. As a result, erstwhile insiders lose their privileged status and join the ranks of unemployed, which insiders clearly dislike. Firms can evidently use this threat of replacement effectively to instil fear among the insiders—in which case the latter, even in presence of LTCs, would not dare to ask for higher wages—due to the very threat of replacement. Now, if insiders cannot indulge in rent-seeking activity and fail to increase their wages above the supply price of labour, then the illustration of unemployment provided above completely loses its meaning. This is a basic lacuna of the I–O theory.

Now authors identify only the firing costs (and not hiring costs) as the reason behind the phenomenon of involuntary unemployment. To demonstrate why hiring costs do not contribute to generating unemployment, the authors bring in another category of workers, namely entrants. Entrants are defined as those “who have recently acquired jobs with a future prospect of gaining insider status, but whose current positions are not associated with significant turnover costs” (Lindbeck and Snower 1988). In plain terms, when an outsider is hired, she becomes an entrant to the firm. Note that since workers are usually recruited through a selection procedure (like advertising, screening, etc), hence some costs are already expended over the entrants. These are precisely the hiring costs that are not expended in similar manner on outsiders. The authors also argue that entrants have “a future prospect of gaining insider status.” This is because normally it takes some time to acquire insiders’ status (referred to in the literature as “initiation period”). Therefore, entrants are yet to achieve insider status so there is no firing cost associated with the retrenchment of entrants. Consequently, entrants (unlike insiders) cannot drive their wages above the reservation wage of the outsiders by both hiring and firing costs.

Nonetheless, entrants can exploit/utilise the hiring costs to drive their wages above the reservation wage of the outsiders, but only by the magnitude of hiring costs. Thus, with three types of workers—insiders, entrants and outsiders—we have a differential wage structure where the insider wage exceeds the entrant wage (by firing cost), which in turn exceeds the reservation wage of outsiders (by hiring cost).

However, the difference between reservation wage of outsiders and entrants wage is not the cause of involuntary unemployment.

The outsiders may be willing to work for a wage that falls short of the insider wage by an amount sufficient to compensate the firms for insider-outsider productivity differences and all production-related turnover costs [ie hiring costs], but outsiders may nevertheless be unable to find jobs. The reason, obviously, is that the rent-related labor turnover costs [ie firing costs], may be sufficiently high to discourage firms from hiring the outsiders. (emphasis added; Lindbeck and Snower 1988)

Note that involuntary unemployment is explained as the firm’s inability to replace insiders by outsiders. This is due to the presence of the costs of firing, since by assumption hiring costs do not pose a problem at all. However, this argument is not correct; later we show that the cost of firing cannot explain the existence of involuntary unemployment.

Next we would look at the diagrammatic representation of the I–O theory that explains involuntary unemployment in terms of firing costs and recommends policy prescription of how such unemployment could be cured. It would be apparent from the following diagram that the basic contours of LMF directly follow from the policy recommendations of I–O theory (for detailed policy discussions see Roychowdhury [2013]).

The basic argument of I–O theory can be demonstrated through the above diagram (Figure 1). In the figure, WErepresents the reservation wage (RW) of the outsiders and AB represents the marginal product curve (MPL) of workers. Now, E represents full employment equilibrium outcome, where the MPL curve cuts the RW of outsiders. Let “h” be the exogenously given marginal hiring cost represented by MN in the figure, and “f” be the exogenously given marginal firing cost represented by NS in the figure. From the preceding discussion, it is clear that if there were only hiring costs then insiders would be able to increase their wage by no more than K (on the MPL curve). Additionally, if there is firing cost as a consequence of JSR, then insiders can manipulate to raise their wages up to J. Also suppose the initial group of insiders be LI in the above figure, such that all insiders are employed when insiders’ wage is OS (that is, insiders’ wage is increased up to J above the RW of outsiders; in terms of notation WI = RW + h + f).

Now without hiring and firing costs profit maximisation would result in the full employment outcome at E. However, with hiring and firing costs LI, pool of insiders would claim wages equivalent to OS (=JLI). Now, even if insiders ask for insider wages above reservation wages up to WI(= RW + h + f)—firm would nonetheless have no incentive to replace insiders by outsiders; for the effective cost of employing an outsider at the margin would be RW + h + f, which is exactly the same as the insiders wage claim. Thus, outsiders cannot underbid the insiders’ wage below OS and it is argued that employment gets stuck at LI below full employment level (LF); where LILF represents involuntary unemployment.

Now it is simply assumed that outsiders are willing to work for less than the insider wage if LTCs involved only hiring costs. However, outsiders are not willing to work for less than insider wage if LTCs include both hiring and firing costs. Thus, in the policy prescription, authors argue that if firing costs are removed then the economy would automatically move towards full employment (LF). Thus, they recommend scrapping of JSR. In terms of the figure this means that if firing cost is scrapped then the economy moves to K (from J) and eventually to E. However, we would show that both assertions are erroneous; we shall demonstrate that (i) full employment is achieved even in the presence of firing cost (assuming, for argument’s sake, that firms do not bear hiring cost); and (ii) hiring costs alone are enough to generate unemployment.

A Critique of the Lindbeck and Snower Argument

In this section we show that the conclusions drawn from the model are not internally consistent. At the outset, note that the entire theoretical literature which identifies JSR as the main obstacle to reach full employment assumes that labour laws extend to all sectors of the economy. In other words, the coverage of JSR is universal. For if this is not the case then, according to their theory, there should not be any obstacle to reach full employment. Only there would be two segments in the economy: a high wage sector governed by JSR alongside low wage sector not governed by JSR. Competition among workers in the non-JSR sector would result in full employment. In other words, I–O theory cannot explain existence of involuntary unemployment, when JSR applies to only a sub-sector of the economy. Thus, to apply a theory which assumes universal JSR to an economy characterised by a large unorganised sector (notably India) is clearly untenable.

Further, it can be shown that even in the presence of firing costs, the representative firm can reach full employment level E. This is the case if we consider a two-tier wage system in the I–O framework. Let us, for argument’s sake assume zero hiring cost. The following is required to attain full employment: let the insiders continue to receive higher wages consistent with firing cost and the representative firm maintains its incumbent workforce. Until now, as full employment is not reached, entrants are employed as new recruits and are paid lower wages (since entrants are yet to attain insider status so no firing costs are associated with entrants).

In terms of Figure 2, consider that the initial incumbent workforce of the firm is LI, where the insiders receive wage WI(=OB) (higher than RW by the magnitude of firing cost and we assumed zero hiring cost). Now, suppose the representative firm does not consider replacement and continues to retain its existing set of insiders (LI) at wage OB, the question that arises is: Is there any opportunity for the firm to add to its existing pool of insiders?

Remember that it takes some time for the newly hired workers to attain insider status. And, it is only after attaining insider status that workers are protected from free dismissal by firing cost. Therefore, new workers can be recruited at the reservation wage (RW) of outsiders (remember we have assumed zero hiring costs for the time being). In that case, profit maximisation would guide the firm to expand employment beyond LI. Actually, in a wage system in which insiders continue to receive insiders’ wage and newly appointed workers, who can be periodically thrown out of employment before attaining insiders’ status, are paid reservation wage—it is easy to see that profit maximisation would allow employment creation up to E and there are no obstacles to full employment. Thus, with a two-tier wage system, in which insiders continue to receive insiders’ wage (OB), while entrants receive RW; there is no obstacle to full employment even in the presence of firing costs. Therefore, the I–O framework is incapable of explaining the existence of involuntary unemployment in terms of LTCs.

Moreover, the I–O theory assumes a perfectly competitive market structure which necessarily implies free entry and exit of firms. In such a situation too the explanation of involuntary unemployment provided by the I–O theory cannot stand scrutiny. The reason is, hitherto unemployed workers can simply come together and start new firms (since there is free entry and exit). Further, since new firms have no insiders’ pool carried over from the past thus they can simply employ outsiders at RW. However, since existing firms continue to pay higher wages to their insiders, therefore, when new firms enter the market, the existing firms would simply be out-competed by the new ones. Under these conditions it is clear that the existence of involuntary unemployment is once again summarily rejected.

Further, for a moment we assume that there are only hiring costs. In terms of Figure 1, insider wage in that case is only ON. Authors argue that in this situation the economy moves to full employment at E. This is because, as the authors argued, outsiders are willing to work for a wage that is sufficiently lower than the insiders’ wage to compensate the firm for all hiring costs. The question arises as to how less the wage of outsiders should be to sufficiently compensate the firm for all hiring costs.

Now if outsiders ask for reservation wage (RW) then the firm would have no incentive to replace entrants by outsiders at the margin, since the effective cost of both types of workers would be the same; consequently the economy would be stuck below full employment at K (on MPL curve). Hence, under such circumstances the hiring cost alone would be sufficient to generate unemployment. But, this is what the authors discard. Thus, to sustain their claim, it must be the case that the outsiders pay for their hiring costs, which means that outsiders accept to work for less than RW by the amount of hiring cost. Only then the entrants would be readily replaced by outsiders and hiring cost alone would be unable to generate unemployment. However, if outsiders are ready to work for less than RW, then this violates the standard definition of RW.

Thus, the I–O theory lands up in a theoretical contradiction in which either hiring costs alone are enough to generate unemployment, thus making LMF inconsequential or, RW loses its meaning (since outsiders accept to work below RW). Thus, the I–O theory (and consequently its policy recommendation of liquidating JSR) is inconsistent even within the structure of that model. Next, we shall examine the effectiveness of carrying out the policy recommendation suggested by the I–O theory when aggregate demand has an autonomous role to play.

Effectiveness of LMF in Face of Autonomous Aggregate Demand

In the previous section we saw that the I–O theory sought to explain persistence of unemployment—in terms of rigid real wages settling above the market-clearing level. This is precisely the method all neoclassical models rely upon for explaining involuntary unemployment.

By now it must be clear that the I–O framework (which is the theoretical foundation underlying LMF strategy) provides a sui generis explanation of wage rigidity—based on insiders’ market power—and consequently argues for LMF, which robs the insiders of their labour market power. Therefore, LMF ultimately wishes to get rid of all LTCs, in order to produce a frictionless labour market, so that real wages can move freely to establish full employment equilibrium. In other words, LMF envisages bringing the real economy to a close approximation of an ideal textbook type neoclassical labour market.

However, the question arises: Even if real life labour markets are made to closely replicate the neoclassical labour market, does it necessarily guarantee full employment equilibrium outcome through real wage flexibility? According to Keynes (2010), this is not so and we shall demonstrate why. It will be further shown that for the Classical16 conclusions (hence predictions of the I–O theory) to hold, it requires strict adherence to Say’s law. Demand has no role to play here.

Now both the I–O theory and the neoclassical theory concentrate only on the labour market for determination of employment. The only difference being that in I–O theory discussions are entirely carried out in the context of a representative firm;17 whereas in the neoclassical approach discussion is carried out in terms of the aggregate labour market. To reach conclusions at the level of macroeconomy in I–O theory, the representative firm is blown up by the number of firms in the economy,18 and it is assumed that a single firm is the microcosm of the whole economy.19

We shall demonstrate below, in the context of the aggregate labour market that even if real wages are made flexible, there can still be persistent unemployment due to insufficiency of aggregate demand in an economy. This is the essence of Keynes’s argument. Keynes argued that it is only under the assumption of Say’s law (of all savings automatically being invested) that real wage flexibility can result in full employment. Let us now turn to Figure 3 (p 45).

The figure depicts the aggregate labour market. The market demand curve for labour is given by the (downward sloping) aggregate marginal product curve for labour (ND), drawn based on diminishing marginal productivity assumption. The labour supply curve is upward sloping (NS) along which the workers equate the utility obtained from the real wages they receive, with the marginal disutility of the amount of labour supplied.

Persistent Unemployment

Full employment is defined at the point of market clearing and is represented by the wage–employment combination W*, N*. Suppose the economy gets stuck at some real wage W1 above the market clearing wage (say, due to trade union activity). According to the neoclassical school, the MPL curve is the labour demand curve. Hence, like a normal demand curve the quantity of labour demand is determined by the independent variable, namely, the real wage rate. In this framework labour demand is determined by the equation ND=f (real wage). Therefore, corresponding to real wage W1, employment is determined at N1 and unemployment is given by N2 - N1.

Now the neoclassical school explains persistence of unemployment by downward rigidity of real wages. In other words, employment remains arrested at N1and the labour market fails to clear because trade union runs a “closed shop.” Therefore, a necessary and sufficient condition for the labour market to clear is to remove such labour market imperfections (by delegitimising trade union activity or removal of LTCs). This allows the real wages to adjust to W* through competition among workers. Let us see what this involves (see Figure 3).

For simplicity, we assume that the economy is characterised by classical savings function, that is, all wages are consumed and all profits are saved. Now, for historical reasons, if real wage is W1,then employment is grounded at N1. The magnitude of total output is given by the area under the marginal product curve to the left of N1, that is, by the area, acde. The wage bill is obtained by multiplying the wage rate with the level of employment and is denoted in the figure by bcde. Since national income is distributed between only two categories of income—wages and profits—profits are represented in the figure by abc.

Now, the nature of capitalist production is commodity production (that is, production for market). Therefore, for a certain amount of output to be actually produced in equilibrium, even in the short period, requires this output to be sold in the market or demanded. Since national output is distributed among wages and profits, therefore we have, Y = W + P; (Y=aggregate income, W=wage bill and P=aggregate profits). Now, let us see how this income is allocated, in the case of a closed economy without government; clearly aggregate income would be allocated between consumption (C) and savings (S)—therefore, Y = C + S. By virtue of classical savings function we have, S = P. If Y is to be an output produced in equilibrium, then it should also be sold in the market and from the expenditure side it must be true that, Y = C + I, where I is aggregate investment expenditure. Therefore, we arrive at the conclusion, I = P.20 Therefore, from Figure 3 investment requirements associated with sustaining N1 level of employment is abc.

With these observations, now suppose that “thoroughgoing competition” is introduced in the labour market, say by scrapping hiring and firing costs, as recommended by the proponents of LMF. Assuming that workers can influence real wages through money wage bargains, suppose that workers successfully bid down real wages to the market clearing level W*. The question arises: Does it necessarily imply that equilibrium employment would increase upto N* (from N1)? From the wage–employment configuration W*, N*, it is clear that the wage bill would be b1c1d1e, which is automatically demanded since all wages are consumed. With total income being ac1d1e, the potential profit or surplus is ac1b1. However, from the above analysis it is clear that this potential profit needs to be realised. Profits need to be absorbed in the economy and the only demand component consists of investment purchases. Hence, aggregate investment must increase exactly by the amount bcc1b1.

However, there is no obvious reason, except by the assumption that all surplus (that is, planned savings) are automatically used for investment purchases, as to why this should be so. This is because investment is autonomously determined and does not simply automatically get jacked up to counter any shortfall in other expenditure components. If investment is not passively adjusting, then accommodating expenditure component equivalent to the extent of bcc1b1is missing from the economy. Hence, a portion of the output (bcc1b1) arising out of expansion of employment from N1to N* goes unsold in the market for which cost has already been incurred. This would entail a loss to the entrepreneur from expansion of employment. Therefore, under the assumption of profit-maximising producers, N* cannot be the new equilibrium position of employment (even when the supply-side conditions justify expansion of employment; essentially because the capitalist nature of production is commodity production), even with real wage flexibility.21 In other words, the above discussion shows that real wages cannot be reduced exogenously, when real wage is determined from the marginal product of labour curve that is, under perfectly competitive conditions.

Neoclassical theory, nonetheless, maintains that real wage flexibility can cure unemployment. Hence, for logical consistency, from the discussion just carried out it must assume that investment would passively adjust to any amount of planned savings automatically, that is, any amount of potential profits must be immediately realised.22 But, such an assumption would mean that demand has no autonomous role to play in an economy and anything produced would always be sold. In other words, this must necessarily assume that “supply of goods creates its own demand” that is, Say’s law of market. Demand has no
independent role to play here and consequently the “multiplier theory” has to be abandoned.

From the above discussion, it is clear that the theories that suggest real wage flexibility (say, by exogenous decree) can get rid of unemployment, assume away the autonomous role played by aggregate demand altogether and must logically bank upon Say’s law of market to sustain their argument. Hence, the I–O theory must be logically necessarily based on Say’s law, while stating that the removal of hiring and firing costs would enhance employment. The logical necessity for such an assumption clearly makes the theory untenable in real life situations.

In Conclusion

The motivation of this paper was to identify the theory underlying the LMF argument in India. To accomplish this objective we reviewed the empirical studies prescribing LMF. From the interpretation of empirical studies it was clear that the authors arguing for LMF were basing their arguments on rigid real wages and explained wage rigidity in terms of LTCs. Such explanations are provided by the I–O theory of employment and unemployment, where involuntary unemployment is explained in terms of the bargaining power of the insiders derived from LTCs. A close scrutiny of the I–O theory revealed that it could not explain the existence of involuntary unemployment, if firms do not bear hiring costs under the following circumstances: (i) a two-tier wage system, and (ii) when new firms enter the market. Moreover, we show that hiring costs alone are sufficient to generate involuntary unemployment (thus, LMF loses its effectiveness) or, if we assume in authors’ line (that hiring costs do not generate unemployment), then the standard definition of RW is violated. Further, we showed that for the policy prescription of the I–O theory to hold good, one needs to necessarily bank upon the operation of Say’s law of market. Since the I–O theory falls short of demonstrating the existence of involuntary unemployment in one case and its policy prescription becomes ineffective in the other (along with its invoking of Say’s law), we, therefore, conclude that the theoretical foundation underlying LMF is unsound.


1 The organised sector comprises of only those establishments which employ at least 10 or more total workers (Indian Labour Year Book 2007: 2).

2 However, it is erroneous to identify the so-called rigid labour laws for the slow growth in employment in the organised sector as a whole. This is due to the domain of application of
the law (Chapter VB of Industrial Disputes Act, 1947) governing lay-off, retrenchment of workers and closure of units. It applies only to organised manufacturing sector, mining and plantations. Thus, restrictions on firing are applicable to only a subset of the overall organised sector. For a discussion on these issues see Roychowdhury (2013).

3 The other major task facing the commission was “to suggest an Umbrella Legislation for ensuring a minimum level of protection to the workers in the unorganised sector” (SNCL
Report 2002: 6).

4 Available theoretical discussions on the subject (namely, Basu [2007] and Basu et al [2009]) assume full employment—this considerably reduces their appeal for an economy openly plagued with involuntary unemployment.

5 Actually, Fallon and Lucas (1991) is an earlier version of Fallon and Lucas (1993).

6 These adjustment costs are in the authors’ words: “On the hiring side certain components of these costs are apparent—the costs of attracting a pool of applicants, the costs of screening the applicants either before appointment or during an initial trial phase, and the costs of training. On the firing side costs may also be incurred. For example, if any worker possessing firm-specific skills quits, this typically imposes a net loss on the firm. Moreover, even without legislated requirements, firms may voluntarily choose to enter a contract with their employees that promises severance pay” (emphasis added) (Fallon and Lucas 1991: 400–01). Clearly, by adjustment costs, the authors mean costs associated with labour entering or exiting a firm, that is, labour turnover costs. Note that JSR may only be instrumental in raising the firing costs but not hiring costs (since stipulation of training costs or, initial trial phase cannot be laid down by law).

7 LTCs operate only through the wage route in raising the cost of labour.

8 It may be thought that there is an element of double counting here, since it is argued that JSR increases LTCs and results in higher wage claims of the currently employed, which raise the cost of employing labour. However, there is no double counting involved here since the argument is: removal of the currently employed workers or insiders involves an expense of LTCs (hiring and firing costs) for firms. Knowing this fact, currently employed workers immediately ask for higher wages (their wage bargaining power determined by the magnitude of LTC), raising the cost of labour. The currently employed workers bid for higher wages, since they understand that the firm would be indifferent between keeping and retrenching them. In plain terms, the currently employed workers demand higher wages, once they understand that firms would not mind paying them higher wages, for, otherwise, in case they are retrenched, firms would anyway have to bear the hiring and firing costs. Thus, LTCs and rising real wages do not increase the cost of labour in additive manner; rather LTCs are instrumental in bestowing bargaining power in the hands of current workers who utilise it to raise their wages.

9 However, expropriation effect is not identified by the authors as an additional factor contributing to adoption of higher capital-intensive techniques, resulting in employment stagnation. Rather, employment slowdown is explained through lower capital stock formation (as investment is discouraged by lowering returns on investments, due to rent-seeking activity of the workers), hence reducing the number of workers required to work on it.

10 At the time of introduction of the labour regulations.

11 Currently, employed are known as insiders and those without employment as outsiders. The adjustment costs typically make labour adjustments costly and hence sluggish. In India, it is claimed that such tendencies are present: “Inertia in employment is consistent with firing restrictions as also with insider activity, for both of which there is evidence in Indian manufacturing” (Bhalotra 1998: 25).

12 The discussion that follows is particularly based on their paper: “Efficiency Wages versus Insiders and Outsiders,” European Economic Review, 31 February 1987: 407–16. However, all references here are from their 1988 book.

13 Wages of existing workers remain rigid since underbidding is not possible, if wage costs and LTCs are equal at the margin, in equilibrium.

14 Rigid wages mean these are inflexible downwards. That it is a sufficient condition for the existence of involuntary unemployment is amply clear from below, “these workers [unemployed] must have an incentive to gain employment by offering to work less than the prevailing wages. In order for involuntary unemployment to be persistent, there must be some other agents who are both motivated and able to prevent such underbidding” (Lindbeck and Snower 1988). In the I–O theory these agents are the insiders.

15 The authors’ notion regarding the instrumentality of real wages in generating unemployment came out most clearly while they were discussing the effectiveness of demand management policies in reducing unemployment. It is best to quote them in some details, “[There] are atleast two routes whereby demand management can raise employment and production: (a) it can raise the price level before the insiders have a chance to adjust their nominal wages, thereby reducing the real product wage and inducing firms to hire more workers, or (b) it can give firms the incentive to expand employment at any given real wage. Route
(a) involves a movement along a downward-sloping labour demand curve, whereas route (b) involves either an outward shift of that curve or a movement along a horizontal segment of the labor demand curve” (Lindbeck and Snower 1988: 266).

16 Keynes (2010) used the term “classical economists” to demarcate Marshall and his followers, the marginalists. However, according to Sweezy (1946: 297), this seemed to be misleading. He argued, “It is preferable to regard John Stuart Mill as the last of the classical economists and to label the Marshallians the ‘neoclassical’ school.” In what follows, we shall demarcate the Marshallians, whom Keynes attacked, as neo-classicalists.

17 Since the I–O theory “aim(s) to provide microeconomic rationales for some workers’ persistent failure to find jobs” (emphasis added) (Lindbeck and Snower 1988: 15).

18 See the methodology adopted to transfer the argument from the level of a single firm to the economy as a whole in the case of Lindbeck and Snower (1988: 70).

19 This method in our view is fundamentally flawed and is fraught with the excesses of “methodological individualism.” For a comprehensive critique of this approach see Patnaik (2008: Chapter 7).

20 The question that arises is as to which way this equation is determined. Kalecki (1971: 78–79), referring to the above equation, brilliantly provides the answer (his analysis assumes that capitalists consume a part of their income): “Does it mean that profits in a given period determine capitalists’ consumption and investment, or the reverse of it? The answer to this question depends on which of these items is directly subject to the decisions of capitalists. Now, it is clear that capitalists may decide to consume and to invest more in a given period than in the preceding one, but they cannot
decide to earn more. It is, therefore, their investment and consumption decisions which determine profits, and not vice versa. Therefore, the equation is determined from the left to the right-hand side; that is, investment is independently determined.

21 This is cogently argued by Burchardt (1944: 6): “The assumption that competitive wage reductions will increase employment presupposes that entrepreneurs will always and immediately react on a temporary increase in actual profits with increased outlay.” He concretely argues, “The crucial point in this theory [neo-classical theory] is the assumption that employers react on the initial increase in profits [following wage reduction] immediately with increased outlay on capital goods or on their own consumption. If they hesitate to do so immediately, employment will not be increased” (p 4).

22 In terms of the equation I = P, this would mean that it is determined from the right to the left hand side, exactly in the opposite direction of what Kalecki (1971) had suggested. It is only under such assumption, it can be logically sustained that investment will exactly offset the excess amount of potential profits generated, associated with the increased employment.


Ahluwalia, Isher Judge (1991): Productivity and Growth in Indian Manufacturing, Delhi: Oxford University Press.

Basu, Kaushik (2007): “Labour Laws and Labour Welfare in the Context of the Indian Experience,” Collected Papers in Theoretical Economics, Volume III, Kaushik Basu, Oxford: Oxford University Press.

Basu, Kaushik, Gary S Fields and Shub Debgupta (2009): “Labor Retrenchment Laws and Their Effect on Wages and Employment: A Theoretical Investigation,” New and Enduring Themes in Development Economics, Bhaskar Dutta, Tridip Ray and E Somanathan (eds), Singapore: World Scientific Publishers.

Besley, Timothy and Robin Burgess (2004): “Can Labour Regulation Hinder Economic Performance? Evidence from India,” The Quarterly Journal of Economics, February, Vol 119, No 1, pp 91–134.

Bhalotra, Sonia (1998): “The Puzzle of Jobless Growth in Indian Manufacturing,” Oxford Bulletin of Economics and Statistics, Vol 60, No 1, pp 5–32.

Burchardt, F A (1944): “The Causes of Unemployment,” The Economics of Full Employment, F A Burchardt (ed), Oxford: Basil Blackwell, pp 1–38.

D’Souza, Errol (2010): “The Employment Effects of Labour Legislation in India: A Critical Essay,” Industrial Relations Journal, Vol 41, No 2, pp 122–35.

Dutta Roy, Sudipta (2004): “Employment Dynamics in Indian Industry: Adjustment Lags and the Impact of Job Security Regulations,” Journal of Development Economics, Vol 73, No 1, pp 233–56.

Fallon, Peter R and Robert E B Lucas (1991): “The Impact of Changes in Job Security Regulations in India and Zimbabwe,” The World Bank Economic Review, Vol 5, No 3, pp 395–413.

— (1993): “Job Security Regulations and the Dynamic Demand for Labor in India and Zimbabwe,” Journal of Development Economics, Vol 40, No 2, pp 241–75.

Ghose, Ajit K (2005): “High Wage-Low Productivity Organised Manufacturing and the Employment Challenge in India,” The Indian Journal of Labour Economics, Vol 48, No 2, pp 232–42.

Government of India (2001): Report of the Task Force on Employment Opportunities, Planning Commission, New Delhi.

Indian Labour Year Book (2007): “Labour Bureau,” Ministry of Labour, Government of India, Shimla.

Kalecki, Michal (1971): Selected Essays on the Dynamics of the Capitalist Economy, Cambridge University Press.

Keynes, J M (2010): The General Theory of Employment, Interest and Money, Delhi: Atlantic

Lindbeck, Assar and Dennis J Snower (1988): The Insider-Outsider Theory of Employment and Unemployment, London: The MIT Press.

Patnaik, Prabhat (2008): The Value of Money,  Delhi: Tulika Books.

Roychowdhury, Anamitra (2013): “Labour Market Flexibility and Conditions of Labour in the Era of Globalisation: The Indian Experience,” unpublished PhD thesis, Jawaharlal Nehru University, New Delhi.

SNCL Report (2002): “Report of the National Commission on Labour, Vols I and II,” Second National Commission on Labour Report, Ministry of Labour, Government of India,

Sweezy, P M (1946): “John Maynard Keynes,” KeynesGeneral Theory: Reports of Three Decades, Robert Lekachman (ed), London: Macmillan, pp 297–304.

World Bank (1989): India: Poverty, Employment and Social Services, Washington, DC.

Updated On : 3rd May, 2019


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top