Lockouts and strikes: some comments on the experience of India and Australia.

Abstract

This paper compares the rise in lockouts relative to strikes in India and Australia. It also compares the overall pattern of industrial disputes in the two countries, noting similarities in the frequency of disputes and differences in the average duration, for example. Although the data upon which comparisons of lockouts are made vary vastly for both countries, and although there are major differences in labour markets and genera/economies, a number of notable similarities nevertheless emerge. These similarities include the relative rise in importance of lockouts, the relatively high duration of lockouts and the shared experience of having government-introduced, market-oriented reforms coinciding with the rise in the relative importance of time lost due to lockouts.

Introduction

Recent research by Briggs (2004a; 2004b; 2004c; 2005) has thrown light on the number of lockouts in Australia during the decade ended 2003. Briggs finds that, among other things, time lost due to lockouts increased during the second half-decade period relative to the first half-decade period. Strikes, on the other hand, decreased over the same time frame. Briggs argues that the relative rise in the importance of lockouts in Australia is attributable to Federal legislative changes during the 1990s that gave employers more power than they had hitherto had. Employers, particularly those operating under federal jurisdiction, responded to the empowering legislative changes of the 1990s by going on the offensive. Not only did lockouts increase, they also tended to be more protracted, on average, than strikes. Thus lockouts accounted for more than half of the working days lost in disputes lasting a month or more during the half-decade ended 2003.

It would be of interest to compare the Australian experience of lockouts with the experience of other countries. One of the great difficulties of analysing lockouts however, is that in most countries national data collecting agencies do not distinguish between strikes and lockouts in their published collections, choosing instead to lump together strikes and lockouts and referring to them collectively as 'industrial disputes' or 'work stoppages'. Even then, when a distinction is made between strikes and lockouts, the data provided can be quite limited, with the disaggregation of data restricted to a small number of broad measures.

India is one country that has, since 1961, distinguished between strikes and lockouts in its official collection of data on industrial disputes. Thus comparisons between Australia's experience of lockouts versus strikes and India's much longer period of reporting on lockouts versus strikes are possible. But there are other reasons why Indian-Australian comparisons are likely to be of interest to labour market observers. First, linkages between the two countries have grown quite strongly in recent years. Australian exports in particular have risen sharply, as has trade in services. As linkages between the two countries expand, it is likely that the two countries will increasingly come to influence one another. Second, the recent rise in the relative size of the Indian economy marks India out as being an influential economic and political power both now and even more so in the future--provided it can maintain its new-found modernising momentum. It is of interest to compare Australian experience with that of this rising power. Finally, both countries have been moving in their own ways towards more market-oriented policies and, in one way or another, moving away from centralised and/or interventionist approaches to governance. This makes both countries of interest in terms of the possible effect that these changes may have on the relative incidence of lockouts and strikes.

Given the above reasons for comparing the experience of Australia and India, this paper will set out to address the following issues. First, how do the overall patterns of time lost due to industrial disputes in India and Australia compare and what factors might explain similarities and differences? Also, how do disputes compare in terms of their frequency, worker involvement levels and duration? Second, what has been the pattern of lockouts versus strikes in India since separate data on lockouts and strikes began being collected in 1961 and what might explain similarities and differences? How do data on lockouts and strikes compare in terms of frequency, worker involvement levels and duration? Thirdly, what evidence is there that legislative and/or administrative 'reforms' have had a differential impact on lockouts versus strikes in India? And finally, can parallels be drawn between Indian experience of lockouts and strikes and Australian experience? Are there any parallels between Australia and India in relation to the effect of employer empowerment on lockouts in absolute and relative terms?

To address these issues, the balance of the paper consists of three main sections. The first gives selected background data on the political economy of both countries. The second section compares and contrasts the broad pattern of industrial disputes in both countries, focussing on the frequency, worker involvement level, duration and overall time lost due to disputes. The third section reviews and compares the experience of both countries with respect to the changing incidence of lockouts. Some concluding comments are offered in the last section of the paper.

Some Background

Trading links between Australia and India have grown quite markedly in recent years. During the year 2003-04, India ranked as Australia's seventh largest export market, accounting for 4.5 per cent of Australia's merchandise exports; just three years earlier India ranked as the thirteenth largest export market (DFAT, 2002, 2004). Figure 1 gives a historical perspective on the recent sharp rise in Australia's export trade to India. Trade in services has similarly risen in importance in recent years. If all of these trends continue, the links between Australia and India will likely widen and deepen considerably in the future.

[FIGURE 1 OMITTED]

Table 1 gives, among other things, a snapshot of some key socio-economic variables for India and Australia. The two countries are markedly different. India is a relatively poor, heavily populated, agrarian society; its per capita GDP is 7 per cent that of the USA, while its total economy is 27 per cent of the size of the USA. Australia is a relatively wealthy, sparsely populated, urban society; its per capita GDP is 79 per cent that of the USA, while its total economy is 5 per cent of the size of the USA. India's growth performance up until 1991 was unimpressive. As Table 1 indicates, per capita GDP growth crawled along at 1.9 per cent per annum on average. That growth rate is reasonable for a wealthy country like Australia, but for a poor country like India, it implies a chronic state of relative poverty.

The United Nations Development Program (UNDP, 2004) estimates a summary index of human development for most of the nations of the world (see UNDP, 2004). Australia is estimated to have the third highest rank in the world, according to this measure. The countries with the highest scores, incidentally, are Norway (rank 1) and Sweden. India's score ranks it at 127. India's human development index is approximately 63 per cent of that of Australia (and the USA), as Table 1 implies. The human development index gives a useful summary measure of the relative development levels of the two economies, indicating as it does the overall balance of differences and similarities in the development of human potential in the two countries.

Table 1 indicates that union density, when measured as a proportion of the organised workforce, has been a little over 20 per cent in India and about 40 per cent for Australia. But as a proportion of the entire labour force (the next row in Table 1), radically different figures apply. Unions have, according to the calculations of Kuruvilla et al. (2002), been much more influential in Australia during the 1980s and 1990s than in, as a standard for comparison, the USA (Table 1). Unions in India, on the other hand, have been much less influential than in either their USA or Australian counterparts. One major factor contributing to the apparent minor status of unions in India is that most employment is in agriculture. Table 1 indicates that agriculture accounts for 60 per cent of employment. Workers in agriculture, especially self-employed workers, have little if any involvement with unions. For workers in the so-called 'formal sector', and especially workers in the government sector, the story is different. Unions are influential within this limited domain. (1)

In fact Kuruvilla et al. (2002) suggest that their index of union influence understates the historical impact of unions in India. They note that: 'indian union density is not that high and its union influence score is extremely low, although Indian unions perhaps have the most supportive institutional environment in Asia ...'. The historical links between unions and political parties, the general thrust of the socialist policies pursued in particular during the first quarter century of India's post-independence years and the legislative support in place to protected formal sector labour and union interests have all resulted in union influence being significantly larger than the Kuruvilla et al. index indicates.

Bhattacherjee (2002, p. 326) notes the paradox of India's low union density but historically noisy expression of labour market turmoil when he points out that:

   Even though union density is very low by international standards,
   India loses more days every year as a result of strikes and
   lockouts than almost any other country [...]. This raises the
   classic question of whether conflict reflects union power or union
   weakness? It certainly indicates that the basic premise of
   industrial pluralism, the regulation of conflict, has not been
   achieved.

India's Political Economy

Given that readers of this journal are likely to be more familiar with Australia's political economy than India's, some further background information of direct and indirect relevance to India's pattern of lockouts and strikes is briefly touched on.

India won independence in 1947. From then until 1964, India's politics were dominated by Prime Minister Nehru of the Congress Party. Nehru pursued socialist policies. The State was heavily involved in the determination of wages and conditions, at least for that small fraction of the labour force employed in the formal sector (Bhattacherjee 1999; Sodhi and Plowman, 2001). After the death of Nehru in 1964, and his successor's death in 1966, Indira Ghandi (Nehru's daughter) became Prime Minister. Indira Ghandi was Prime Minister from 1966 to 1977 and from 1980 to 1984. She was assassinated in 1984. While a number of reforms were introduced by Indira, by and large she had little success in dealing with the problems of poverty and low growth. Her most controversial actions occurred in 1975, when she declared a national emergency, giving her the power to silence and/or imprison her political opponents. She also banned strikes which, as we shall see, resulted in a sharp reduction of strikes during 1976, though lockouts were largely unaffected. In the election of 1977 she lost both government and her own parliamentary seat, though she was subsequently voted back into office in 1980.

Indira Ghandi's son, Rajiv Ghandi, was appointed Prime Minister after Indira's assassination. His administration (1984-89)introduced some reforms, 'partial liberalisation' according to Zagha (1999, p. 171), but these reforms were insufficient to address the underlying malaise. The underlying malaise of slow growth and entrenched poverty, according to most commentators, was attributable to a system of production that was inflexible, over-regulated and grossly inefficient. The extraordinary regulatory environment, intended presumably to protect workers and prevent the rise of monopoly capitalism, is well documented in Bhagwati (1993), Zagha (1999), Davis (2000), Venkata Ratnam (2001), Sodhi and Plowman (2001) Bhattacherjee (2002), Krueger and Chinoy (2002) and The Economist (1997, 2005), amongst others. Some of its features included:

* Job security regulations that mae it near-impossible for firms employing 100 people or more to retrench workers without government approval, which was rarely given (Zagha 1999, p. 165)

* Legislation requiring large firms to offer employees '... benefits such as pension funds, health insurance, creches, subsidised canteens and clinics. The bigger the firm, the more demanding the requirements'. (The Economist 1997, p. 16)

* An inconvertible currency that interfered with international trade and restricted the importation of overseas technology

* High tariff protection (87 per cent in 1990-91 : Venkata Ratnam, 2001, p.139) further preventing international trade and specialisation and its concomitant benefits

The above are merely indicative of the regulations in place intended to protect the Indian economy and Indian formal-sector workers from the outside world. The effect of the regulatory labyrinth was to prevent firms from growing to develop and exploit economies of scale, stunt productivity growth, limit the growth of a strong formal sector (Krueger and Chinoy, 2002) and in the long run perpetuate relatively low average living standards.

The above problems were not effectively addressed during Rajiv Ghandi's period in office, although limited reforms were introduced. These reforms included a long-overdue recognition that higher productivity was necessary to underwrite higher real wages (Sodhi and Plowman 2001, p. 56) and a partial liberalisation of international trade which gave firms easier access to overseas inputs and technology (Zagha 1999, p. 167). But the reforms were insufficient to avert a crisis in 1991, at which time India had insufficient foreign exchange to pay for its overseas loans.

The Congress Party government, led by Narasimha Rao (Rajiv Ghandi was assassinated in May 1991), decided to follow an IMF-World Bank rescue package involving the depreciation of the currency and the introduction of a number of market-oriented reforms. The introduction of reforms in June 1991 was a radical departure from past practice. Davis (2000, p.65), drawing in part on Bhagwati (1993), observes that: 'If the reform program adopted by Indira Ghandi were 'reform by stealth', and those by Rajiv Ghandi "reform by reluctance", then those by Prime Minister Narasimha Rao and his Finance Minister Manmohan Singh following the "1991 Crisis" were "reform by storm".' Since the reforms of 1991, per capita GDP growth has averaged 4.2 per cent per annum (Table 1).

While some the market-deregulation reforms since 1991 have been quite far reaching, such as currency and tariff reforms, formal labour markets deregulation has been minimal. The Economist (1997, p. 19) notes that: '... except for the indirect effect of increased competition, the reforms of 1991-93 left the labour market untouched'. Similarly, in a more recent comparative survey of India and China, The Economist (2005, p. 14) again makes the point that:

   Firing workers remain hard in India, which deters firms from hiring
   them in the first place. This has a big impact on productivity.
   Many firms feel unable to reap economies of scale, because once
   they employ more than 100 people they become subject to restrictive
   labour laws.

Nevertheless, there has been a discernible change in the balance of power between unions and employers. But this has not been brought about by milestone legislative changes. Instead, a combination of government administration rulings and actions, including certain individual state government actions (Kuruvilla et al, 2002; Beasley and Burgess, 2002; Sanyal and Menon, 2004), in conjunction with a number of judicial rulings against unions (Sharma, 2004), have been responsible for the reforms that have thus far materialised. Sharma (2004, p. 11) observes that:

   Unable to initiate legislative action because of the constraint of
   parliamentary democracy, the state has resorted to executive action
   for reinterpreting legal rights of labour in the light of the
   prevalent values. The Parliament has not so far been able to
   liberalise labour legislation to a significant extent despite the
   recommendations of the second National Commission on Labour. (2)

Broadly then, there are three periods of particular interest when examining the pattern of industrial disputes in India. The first is around 1975-1976 when Indira Ghandi temporarily outlawed strikes during her declared 'national emergency'. The second period is from around 1985, when Rajiv Ghandi's administration introduced 'partial liberalisation'. The third and by far the most important period is from 1991. We will be particularly interested to see how the pattern of lockouts and strikes have changed during these three periods.

Industrial Disputes: Broad Comparisons

To get an overview of the pattern of industrial disputes in India and Australia, we compare 'time lost' due to industrial disputes for both countries. Time lost is here initially defined as the number of working days lost due to industrial disputes per member of the labour force. (3) Note the denominator (the labour force) chosen here is a little unconventional. Time lost per employee or time lost per potential employee (employees plus the unemployed) or perhaps time lost per unionist are probably more conventional measures.

We have chosen the labour force as the denominator for 'time lost' on the following grounds. Given the relatively small size of the so-called formal sector and the accompanying dominance of agriculture as a source of employment, applying, say, the number of employees as the denominator significantly magnifies the dimensions of industrial disputes relative to the overall size of the total economy. According to ILO internal data, formal sector employees, during the 1980s and 90s, averaged a little over 7 per cent of the entire labour force. Unionists averaged a little under 2 percent of the entire labour force. Thus, applying the labour force as the divisor for 'time lost' places the Indian data into a broader context. Arguably, this broader-context measure is a more accurate measure of the economy-wide significance of disputes than an alternative measure that applies formal sector employees or unionists as a divisor. (4)

Figure 2 charts time lost (per labour force member) for both countries. Note the scale of the vertical axes are different for the two countries. Some features of the figure include the following. First, in spite of the many differences in the socioeconomic backgrounds of the two economies, there are some broad similarities in the pattern of disputes. For the timeframe considered, both countries experienced initially rising time-lost rates, followed by declining rates. The correlation coefficient for the two series is 0.43. (5) Second, whereas the Australian data peaked during the mid-1970s, the Indian data did not peak until the early 1980s. Nevertheless, both countries have experienced broadly downwards trends in their data after that time. Third, Australian time lost has, on average, been higher than in India, particularly during the 1960s and 1970s. Over the entire timeframe, Indian time lost averaged out at about half that for Australia. However, in more recent years (since, say, 1991 the year in which major economic reforms were introduced in India) time lost in India has averaged out at about the same as time lost in Australia.

[FIGURE 2 OMITTED]

What lies behind these differences and similarities?

If industrial disputes in both countries were responding in a similar fashion to determinants such as inflation, union density and labour demand, then an analysis of these different determinants might provide an answer to the question posed above. As we shall see, however, it is unlikely that India and Australia respond in the same fashion to such determinants. Moreover, there are different views as to exactly what the determinants are, especially in reference to recent Australian analyses.

In the Australian case, there is a considerable literature on the determinants of industrial disputes. Recent econometric-based analyses have been advanced by Chapman and Gruen (1991), Morris and Wilson (1999), Perry and Wilson (2001) and Hodgkinson and Perera (2004). Recent discussion-based analyses have been advanced by Sodhi and Plowman (2001), Healy (2002, 2004), Perry (2004) and Briggs (2004), among others. These analyses are not unanimous in their findings and/or emphases. On balance it is suggested the following determinants warrant further consideration within the context of the purposes of this paper; these determinants are inflation, changing union density, legislative/ administrative changes, worldwide developments and underlying labour market excess demand conditions. (6) Note that most recent studies have down played underlying labour market conditions, as reflected in the local unemployment rate, as a factor explaining changes in Australian industrial disputes (Morris and Wilson 1999; Healy, 2002; Hodgkinson and Perera, 2004). One reason for this is the simultaneous decline in industrial disputes and the unemployment over the last decade or so. Such simultaneous declines are contrary to what most models of industrial disputes and the business cycle predict (e.g. Rees, 1952).

In the case of India, there appear to be no econometric analyses to draw on for the purpose of identifying possible significant determinants of Indian industrial disputes. Research on India has focussed almost exclusively on the effect of legislative/administrative changes, (7) though there has been some recognition of the effect of labour market excess demand conditions, at least for the 1960s and early 1970s. (8)

Let us then briefly compare and contrast Indian and Australian experience of (i) inflation, (ii) union density, (iii) legislative/administrative changes, (iv) worldwide changes in disputes and (v) unemployment rate changes so as to get a measure of the extent to which these possible determinants of industrial disputes may have contributed to differences and similarities in the pattern of disputes in the two countries.

First inflation, though important to most Australian analyses, has not been directly suggested to be a systematic contributing factor to changes in Indian industrial disputes. One reason for this may be that, whereas there is a strong one-on-one correlation between annual inflation and annual time lost due to industrial disputes for much of the last half century for Australia, the same does not apply to India. The simple correlation coefficient for annual time lost and annual inflation for Australia is 0.73; for India it is 0.30. Furthermore, while there is some similarity in the spiking pattern of inflation and time lost due to industrial disputes in India for some years such as 1973 and 1979, for other years, such as 1982 when time lost reached an all time high, there is no corresponding peak in inflation.

It might be argued that 1982 was an aberrant year. It was, after all, the year of the Bombay Textile Workers Strike--India's longest strike since independence (Bhattacherjee, 2001) and claimed to be perhaps the world's greatest ever strike (Bidwai 1997). (9) Even the exclusion of this possibly aberrant year however, still generates a relatively low correlation coefficient of 0.36, which rather suggests that annual inflation is not as simply and consistently related to annual industrial disputes in India as it has been, to date, in Australia.

Second, union density, though important to most Australian analyses (cf. Hodgkinson and Perera, 2004), has not been directly suggested to be a contributing factor to changes in Indian industrial disputes. For Australia there has been a well-documented decline in union density since the early 1980s. But this has not been India's experience. (10) So far as can be made out, Indian union density shows no direct relation to time lost due to industrial disputes. For example, union density data for India have not fallen consistently over the last couple of decades, unlike the Australian data. Instead, Indian density rates have undulated over the same timeframe that Australian density figures have been falling. Indian density fell during the late 1980s until 1993, rose between 1993 and 1997 and then declined again between 1997 and 2000. Perhaps the more pronounced decline in industrial disputes in Australia than in India can be attributed in part to the pronounced decline in union density in Australia compared to India's undulating density changes.

A third area of influence on industrial disputes concerns legislative/administrative changes. Both countries have moved pretty much in the same direction; in the direction of market liberalisation. These changes have been gradual and cumulative in the case of Australia, commencing perhaps in the 1970s (Sodhi and Plowman 2001) and gaining momentum during the 1980s and 1990s with, in particular, the installation of enterprise bargaining and the implementation of the Industrial Relations Reform Act 1993 which enshrined into law the right to strike and the right of employers to lockout employees under certain circumstances.

Changes have also been quite gradual in India--at least up until 1991. Though the changes from 1991 appear more dramatic than the changes witnessed in Australia, the fact that Indian labour market changes have not been buttressed by major labour market legislative changes (as discussed in the previous section), has meant that reforms in India have been more tentative than might otherwise have been the case.

Regarding Australia, it has been argued, particularly by Chapman and Gruen (1991), Morris and Wilson (1999), Healy (2002) and Hodgkinson and Perera (2004), that various legislative/administrative arrangements have had a major --even defining--impact on industrial disputes. (11) Similarly for India the (partial) movement away from central government planning and control has been identified as being a major contributing factor to the decline in strikes (as a major component of industrial disputes), and the relative rise in importance of lockouts (Sen Gupta and Sett 2000; Sodhi and Plowman 2001 ; Venkata Ratnam 2001 ; Bhattacherjee 2002 and Datt 2003a, 2003b). We discuss Indian lockouts and strikes in more detail in the next section of this paper.

A fourth area of influence on Indian and Australian industrial disputes concerns the general worldwide decline in industrial disputes. Just as major changes in, say, inflation, unemployment, managerial practices, technology (e.g. computer technology) and fashions in everything from architecture and design to music and clothing get transmitted worldwide, so major changes in the pattern of industrial disputes appear also to be transmitted worldwide. Over the last couple of decades, most of the world's developed and leading economies, such as the USA, Japan, Germany, France, Italy and the UK, have experienced a declining trend in time lost due to industrial disputes. Arguably these trends may have influenced, to a greater or lesser extent, the incidence of disputes in both India and Australia. The process may involve, among other things, an international demonstration effect that we might expect to become more powerful as economies globalise and interdependence increases. It may also be the case that all countries are, in part, responding to common global developments, as well as responding to one another (Perry and Wilson 2001, 2003). (12)

A fifth area of potential influence on Indian and Australian industrial disputes concerns general labour market conditions as reflected in, say, the unemployment rate. (13) It has already been noted that labour market conditions in Australia, as reflected in the Australian unemployment rate, have not in more recent studies been linked to declining industrial disputes. This may not be the case for India, however. India's relatively high unemployment rates from around the mid 1980s may have acted as a moderating influence on strikes (though not lockouts).

On the other hand, during the 1970s and early 1980s when India's industrial disputes (including strikes) where rising most markedly, so too was India's unemployment rate. Perhaps formal sector employees were largely insulated from the deteriorating labour market of the 1970s and 1980s, at least up until the mid-1980s, because of the previously-discussed job security regulations in place. With the partial deregulation of the mid-1980s and the more substantial reforms of 1991, labour market conditions may have become a more pertinent consideration to would-be strikers, and these developments may have acted to moderate strikes. (14)

While there are some notable similarities in the pattern of time lost due to industrial disputes for India and Australia, it is important to keep in mind that there are also major differences.

For a start, lockouts historically have been much more important in India than in Australia. For the period under review, lockouts in Australia have likely been a recent phenomenon. For India, lockouts over the last two decades have dominated industrial disputes. Also, India's industrial disputes and union activity have been largely confined to its formal sector and this sector accounts for less than 10 per cent of all employment (as noted earlier). This highlights the point that when India's industrial disputes are expressed as a ratio of formal sector employees, this puts India's dispute rate amongst the highest in the world.

To sum up, it is suggested that the similarity in time lost due to industrial disputes (and perhaps more so, time lost due to strikes) in both countries has probably been linked most strongly to (i) similar (in spirit if not in detail) deregulatory administrative/legislative changes over the last approximate two decades, (ii) similar sensitivities to worldwide changes in industrial disputes--though perhaps more relevant for more recent years for India as its exposure to the outside world has grown and (iii) often similar experiences with respect to inflation and a sensitivity on the part of both countries to inflation changes. Nevertheless, it seems likely that Australian time lost due to industrial disputes has been more consistently and predictably sensitive to inflation than has been the case for India.

Structural Composition of Industrial Disputes

The information in Figure 2, i.e. time lost due to industrial disputes, can be decomposed into the measures of 'frequency', 'involvement' and 'duration' as below:

Time Lost [] Frequency x Involvement x Duration

WDL/LF [] NID/LF x WI/NID x WDL/WI

In the above decomposition, WDL refers to working days lost due to industrial disputes, LF refers to the labour force, NID refers to the number of industrial disputes and WI refers to the number of workers involved in industrial disputes. Both India and Australia collect and publish these data, though the methods of collection do vary somewhat between the two countries (see Perry and Wilson 2004). We shall employ this 'break-up' of time lost, commencing with the frequency of disputes (the number of disputes as a proportion of the labour force).

Figure 3 charts the frequency of disputes for India and Australia. Note that the scales of the vertical axes are different for the two countries. Two observations are offered. First, the two series are highly correlated, with a correlation coefficient 0.88. For both countries the frequency of disputes peaked during the 1970s--a period of relatively high inflation in Australia, India and elsewhere. (15) After the early 1970s, the frequency of disputes, more or less, trended down. Second, on average, the frequency of disputes for India is about 2.5 per cent that of Australia. Thus the frequency of industrial disputes relative to the labour force has been much more pronounced in Australia than in India.

[FIGURE 3 OMITTED]

Figure 4 charts the average level of worker involvement in disputes. Up until the new millennium, the level of involvement in disputes for both countries was remarkably similar. The average number of workers involved in disputes during the 1960s and 1970s was around 600 for India and around 500 for Australia. During the 1980s and 1990s the respective averages were around 800 and 600, suggesting a possible mild upward trend over the course of the decades for both countries. As the figure indicates, during the new millennium, India's involvement rate rose sharply, while Australia's subsided.

[FIGURE 4 OMITTED]

Figure 5 looks at the duration of disputes. Here there are some marked differences. Indian disputes are, on average, much longer than Australian disputes. If we split the 43 years of observations into approximate halves, the average duration of disputes in India, during the first half (1961-82) was around 14 days, whereas for Australia the average was a little over 2 days. During the second half (1982-2003), the average duration for India is 24 days compared to a little under 2 days for Australia. The broad trend has been for the duration of disputes to rise somewhat in India and decline a little in Australia. (16)

[FIGURE 5 OMITTED]

To sum up this brief comparative overview of the structural composition of disputes in India and Australia, we can note that there are some interesting differences and similarities in the broad pattern of disputes in the two countries. Indian disputes tend to be relatively drawn out affairs, whereas Australian disputes are of a relatively short duration. The number of disputes in India is not, on average, as large as the number of disputes per labour force members in Australia, but the disputes, particularly the lockouts (as we shall see in the next section), do in a sense make up for this. To give an Australian perspective on the relative duration of disputes in India, the average duration of disputes in India during the 1960s and 1970s is comparable to the average duration of the punishing, 'trials of strength' disputes of the pre-World War II period in Australia. The average duration of Indian disputes during the 1980s, 1990s and beyond, on the other hand, exceeds on average the duration of the Australian pre-War disputes by around 80 per cent.

The similarities between the structural composition of disputes in the two countries are to be found in the broad pattern of the frequency of disputes and, to a lesser extent, the overall time lost due to disputes. Also, for many years, up until the new millennium, the average levels of involvement in disputes were very similar; however in recent years paths have diverged markedly. Focussing on the last couple of decades or so, both countries have experienced declining dispute numbers and time lost. The decline in time lost for Australia however has been somewhat more precipitous than the decline for India.

Lockouts in India and Australia

In this section we seek to compare India's experience of lockouts with that of Australia. Official Indian data on lockouts stretch back to 1961. On the other hand, there are no official collections of lockouts data for Australia. An unofficial collection by Briggs (2004a) is all that is 'available'. Notwithstanding these differences, it will be argued that the data reveal a number of interesting parallels in the experience of both countries.

Lockouts in India

We commence with India, Figure 6 displays the overall level of time lost (per labour force member) due to (i) lockouts and (ii) strikes. Three features are noted. First, there has been a general increase in the relative importance of lockouts over time. Since 1987 lockouts have consistently accounted for more time lost than strikes. Second, time lost due to strikes has been more volatile than time lost due to lockouts. And thirdly, time lost due to strikes and lockouts have both separately trended up until 1982 and trended down thereafter. The trend is more pronounced for strikes than for lockouts however, and the downward trend for lockouts between 1982 and 2003 is quite weak.

[FIGURE 6 OMITTED]

What might explain this pattern?

First, regarding the relative rise in lockouts, most commentators argue that this is attributable to a gradual change in the balance of power away from employees and trade unions towards employers and management (Sen Gupta and Sett 2000; Bhattacherjee 2001; Datt 2003a,b).

The relative rise of lockouts commenced around the mid-1970s, leaving aside the effect of the suspension of strikes during the 1975-1976 'National Emergency'. Sen Gupta and Sett (2000) argue the changes were initially brought about by 'economic stagnation throughout the seventies' (p. 147). Depressed demand put pressure on industrialists to restructure and to cut labour costs--both real wage rates and employment levels. Given the legal difficulties involved in retrenching workers, employers used various devices to cut costs. These included 'casualising' the workforce, subcontracting work, defaulting on loans and accounts-due so as to force a closure of certain operations, and hiring labour informally (so as to circumvent government reporting requirements). But, the most frequently employed method of cutting labour costs, according to Sen Gupta and Sett (2000), was through lockouts. Lockouts put pressure on workers to moderate wages claims, they reduce employment levels when product demand is low, and they allow employers to avoid closing down operations altogether, though lockouts can be used to facilitate full and partial closures as well (Mathur, 1991; Sen Gupta and Sett, 2000).

The continued rise in the relative importance of lockouts can be linked to the gradual movement towards market liberalising policies--particularly during the mid 1980s and even more so with the reforms of 1991. And although labour market legislative change has been minimal (as noted earlier), the cumulative effect of both state and central government discretionary rulings in conjunction with a number of judicial rulings has had the effect of further strengthening the hand of capital against that of labour (Datt 2003a,b).

Second, regarding the relative stability of time lost due to lockouts versus strikes, it is hypothesised that strikes are likely to be somewhat more sensitive to price changes than would be the case for lockouts. The reason for this is that an increase in inflation (see Table 1), in the absence of automatic comparable money wage rate adjustments, is likely sooner or later to result in increased worker and union agitation, including increased strike activity, designed to extract money wage rate increases to restore real wage rates. In addition, divergent money wage adjustments are likely to affect wage relativities, which may lead to further union agitation. From the point of view of employers, on the other hand, an increase in inflation will, in the absence of compensating money wage changes, lead to a welcome reduction in real wage costs. The real wage cost reduction will be achieved 'passively', in the sense that it will not involve any action (e.g. lockouts) on the part of employers--just inaction in regards to increasing wages. (17)

Third, regarding similar broad rises in lockouts and strikes up until 1982 and the subsequent decline (though a much weaker decline for lockouts), it is hypothesised that perhaps a mix of factors may have been of relevance. The rise up until 1982 in time lost due to lockouts may be linked to a general weakening of labour market conditions during the 1970s and 1980s. This was reflected in the quite sizable rise in the unemployment rate during these years. Also, the fact that there was a higher level of strikes and thus grounds for antagonism between capital and labour may have helped increase the probability of lockouts occurring. Finally, there may have been an international demonstration effect influencing lockouts and strikes. The worldwide wave of industrial disputes during and around the 1970s may have acted to encourage and validate similar actions in India. Note however that the worldwide increase and subsequent protracted decline in disputes involved strikes much more than lockouts. Thus the international influence on Indian lockouts would have been much more tempered than the international influence on Indian strikes.

To get a perspective on the structure of strikes and lockouts in India, let us decompose lockouts and strikes into the earlier-discussed dimensions of (i) frequency, (ii) involvement and (iii) duration.

Figure 7 charts the frequency of strikes and the frequency of lockouts in India. It is notable that, among other things, there has been a sizeable reduction in the frequency of strikes. The frequency of lockouts has been relatively stable, though during the post-1991 period, the frequency of lockouts has also trended down--albeit in a much less dramatic fashion. Figure 8 indicates that the numbers involved in strikes and lockouts are fairly similar, although involvement in strikes has increased somewhat more rapidly than the involvement in lockouts. There has also been a recent remarkably sharp increase in the average number of workers involved in both lockouts and particularly strikes, though at this point in time it is unclear as to how permanent or otherwise these changes will turn out to be.

[FIGURES 7-8 OMITTED]

Figure 9 indicates that the duration of lockouts is substantially higher than the duration of strikes. On average, between 1961 and 2003 lockouts lasted 49 days, while strikes lasted 11 days. Figure 9 illustrates that the duration of lockouts rose quite markedly between 1961 and 1990. If we divide the last four decades or so into three eras, following Datt's chronology, (18) we get the following readings for the average duration of lockouts:

1961-75 = 28 days 1975-90 = 68 days 1991-03 = 50 days

[FIGURE 9 OMITTED]

The coresponding readings for the average duration of strikes are:

1961-75 = 8 days 1975-90 = 14 days 1991-03 = 11 days

These data indicate that there has been a tendency for the duration of both strikes and lockouts to increase over time, although more recent experience has seen a slight moderation in duration. Also notable is the average duration of lockouts in India is, since the mid-1970s, around four or five times that of strikes.

To get a perspective on the sources of change in the rise of time lost due to lockouts compared to strikes, we can look at broad trends in the relative frequency, involvement and duration of lockouts to strikes. Figure 10 does this. It charts the ratio of lockouts to strikes in terms of frequency, involvement and duration. To smooth the data so as to facilitate long-term comparisons, original series are converted into 5 year moving averages. The data are then converted into indices with base year values set at 100 for 1965. Note also that, given that ratios are being compared over time, a log scale is employed on the vertical axis.

[FIGURE 10 OMITTED]

From Figure 10, it is clear that the principal driver of rising time lost due to lockouts (relative to strikes) is the relative frequency of lockouts. Essentially, the strong decline in the frequency of strikes (Figure 7) has far outstripped the comparatively modest decline in the frequency of lockouts. The relative duration of lockouts has risen and fallen over the years, and although there has been a recent rise in relative duration, there has been no evidence of a consistent or strong underlying upward trend over the last few decades. Relative involvement, on the other hand, has trended down, suggesting that this measure has not been a factor contributing to the relative rise in time lost due to lockouts in India.

Finally, how do lockouts and strikes compare for different Indian industry categories? In Briggs'(2004a) study of Australian lockouts (to be discussed below), it was found that lockouts were predominantly in the manufacturing sector What of India?

There appear to have been no analyses to date of long-term aggregate time series data on lockouts versus strikes classified by industrial sector. Some state-wise analyses have been carried out however. Datt (2003a) compares trends in lockouts based on state government data for West Bengal. He notes that aggregate state data are inconsistent with aggregate nationally-compiled data for West Bengal. Notwithstanding this anomaly, the industry sector for West Bengal is divided into six sectors: Jute, Engineering (major), Engineering (minor), Cotton, Tea and Other. Of these classifications, Engineering (major plus minor) probably comes closest to the Australian classification of 'manufacturing'. Datt (2003a) compares the six years before reform (1986-1990) with the six years after reform (1991-1997). Lockouts in Engineering accounted for 17 per cent of the total during the first period and 20 per cent during the second period. As it turned out, all sectors experienced an absolute decline in lockouts during the second period relative to the first. The smallest proportional decline was in Engineering, wherein lockouts fell by 33 per cent, compared to an overall decline in lockouts of 43 per cent. All in all, it is not possible, on the basis of the very limited data available, to make any definitive comments about the overall impact of reform on the relative impact of lockouts on manufacturing versus other industries. But on the basis of what little information is available, the phenomenon of lockouts being mainly confined to the manufacturing sector does not appear to apply to India.

Lockouts in Australia

As mentioned earlier, information on lockouts in Australia is limited. It is confined, by and large, to the period 1994-2003. Briggs (2004a) believes that lockouts were pretty much 'unheard of outside of the ... 1890s and Great Depression [of the 1930s]' (p. 111).

Briggs points out that in more recent years lockouts were given a firmer legal status as a result of certain legislative changes. The first was the Industrial Relations Reform Act (1993). Under this legislation, lockouts were given limited immunity from common law sanctions against breach of contract. These changes were introduced by the government of that time to permit employer-employee 'balance' in arriving at enterprise bargaining agreements.

The second legislative change was in the Workplace Relations Act (1996). This legislation' ... retained the key elements of the lockout provisions of the 1993 Act and extended the scope of lockouts, notably by introducing the right to use them as a means of pressuring employees to sign individual, Australian Workplace Agreements' (Briggs, 2004a, p. 102).

As a consequence of these changes, the relative bargaining position of employers has strengthened. Apparently emboldened by the legislative changes plus other developments, employers have acted more aggressively in using lockouts as a means of defending their interests, according to Briggs. Thus for the half decade ended 2003, lockouts accounted for 9.3 per cent of all working days lost in Australia; in the preceding half decade they accounted for 1.6 per cent of days lost. Lockouts have been particularly prevalent in the manufacturing sector. During the half decade ended 2003, they accounted for 27 per cent of days lost in manufacturing, compared to 3 per cent in the preceding half decade. Lockouts have tended to be of a longer duration than strikes. During the half decade ended 2003, they accounted for 58 per cent of disputes lasting one month or more; whereas in the preceding half decade they accounted 8 per cent of disputes lasting one month or more.

The broad structure of strikes and lockouts in Australia for the decade ended 2003 is indicated in Table 2 (see Perry, 2005). Note the sharp rise in lockouts in all industries (323 per cent) during the half-decade ended 2003, especially in manufacturing (940 per cent). Interestingly, there was, over the same timeframe, a decline in lockouts in non-manufacturing industries. (19) Whereas there has been an overall rise in lockouts, there has been an overall decline in strikes over the same timeframe. The decline in strikes is applicable to both the manufacturing and non-manufacturing sectors.

Indian-Australian Commonalities

From our brief overview of strikes and lockouts in India and Australia, a number of interesting similarities can be noted.

First, both countries have experienced a relative rise in time lost due to lockouts over the last decade. In the case of India, a relative rise in lockouts has been occurring over most of the last four decades. In the case of Australia, it seems unlikely that lockouts were in anyway significant up until the last decade, though strictly speaking there are no data on the issue to verify the suggestion.

Second, in both countries legislative and/or administrative changes have been associated with the greater relative amount of time lost due to lockouts. In the case of India, these have been mainly administrative changes, as legislative changes have been difficult to implement. In the case of Australia, the introduction of the Industrial Relations Reform Act (1993) and the Workplace Relations Act (1996) have given greater scope to employers to employ the lockout in negotiations.

Third, the average duration of lockouts in both countries is markedly higher than the duration of strikes. In the case of India, the duration of lockouts has, over the last approximate three decades, been on average around four or five times greater than the duration of strikes. In the case of Australia, the duration of lockouts can be inferred to be greater than the duration of strikes, because a preponderance of lockouts have occurred for periods of one month or more, which by Australian standards is quite a long time.

Fourth, both India and Australia have experienced declining trends in time lost due to strikes, while time lost due to lockouts has increased over the last half-decade. In the case of India, there has been an upward trend in the data from around 1997 to 2003. In the case of Australia, there has been an increase in lockouts during the half-decade ended 2003, relative to the half-decade ended 1998.

Lastly, the relative rise in lockouts in both countries has occurred within a backdrop of government-introduced, market-oriented reforms. In the case of India, the reforms were mainly introduced as a result of the economic crisis of 1991. While there were some reforms in the post-Emergency period under Indira and Rajv Ghandi, these changes in no way compared to the major changes introduced in 1991. In the case of Australia, the reforms can arguably be seen as a part of an ongoing process introduced during the 1970s, further developed during the Hawke-Keating Labor government years (1983-96) and continued during the Howard Coalition Government years (1996 and after).

Some Concluding Thoughts

This paper has set out to compare Australian and Indian experience of strikes and lockouts, with a focus on the changing relative importance of lockouts in both countries. India is one of a small number of countries that collects, on an official basis, separate quite detailed data on lockouts and strikes, so it is of some interest to compare India's experience with that of Australia.

As it turns out, there have been some notable similarities in the pattern of strikes and lockouts in the two countries. Both countries have had fairly similar overall patterns of industrial disputes, particularly with regards to the frequency of disputes and time lost due to disputes, over the last four decades or so. Both countries have experienced an increase in time lost due to lockouts relative to strikes, within a context of generally declining overall dispute levels. Both countries find that lockouts, on average, last much longer than strikes. Both countries have had, over the last couple of decades, quite similar experiences of declining industrial disputes like that of most other major world economies. And both countries have, in their own ways, introduced a number of market-liberalising reforms that have had the overall apparent effect of weakening, ceteris paribus, the relative bargaining position of employees and unions and been thus causally linked to the relative rise in time lost due to lockouts.

Nevertheless there are some notable differences as well. Indian disputes, both strikes and (especially) lockouts, are of a much longer duration on average than is the case for Australia. Lockouts in India are a much more important source of disputes in India than is the case in Australia. In Australia lockouts have accounted for a small fraction (about 10 per cent in recent years) of total time lost due to recent industrial disputes, whereas in India lockouts have accounted for more than 50 per cent of all disputes since the mid-1980s. India's industrial disputes are largely confined to its relatively small formal sector, whereas in Australia such a formal versus informal sector dichotomy does not exist. Within India's formal sector, time lost due to strikes and lockouts is amongst the highest in the world, certainly much higher in terms of time lost than is the case for Australia. It is only because this sector accounts for such a small fraction (less than 10 per cent) of the labour force that its economy-wide impact is so attenuated.

Our review of the experience of both countries suggests that there are some notable parallels between Australia and India in relation to the effect of employer empowerment on lockouts in absolute and relative terms. Lockouts, not surprisingly, are more likely to occur when employers are empowered to employ lockouts as a bargaining tool. This might be due to legislative changes or government administrative decrees that sanction in one form or another such actions. Lockouts are more likely to occur when firms are in a weakened financial position and forced to downsize operations (though the evidence for Australia is somewhat more circumstantial). And lockouts are more likely to be longer lasting than strikes.

Finally, a word of caution: it should be reiterated and thus emphasised that our observations and comparisons regarding lockouts in Australia and India are based on quite different data sets. The Australian data are limited to the decade ended 2003 and they incorporate data that have been unofficially compiled. Indian data, on the other hand, are officially compiled, are quite detailed and stretch back to 1961. As a result of these differences, the conclusions drawn from much of the comparative analysis of lockouts need to be seen as being more than usually tentative and limited. It would, nevertheless, be of interest to compare these finding for Australia and India with the experience of other countries. Unfortunately, international data availability and collecting difficulties are likely to inhibit, if not prohibit, such an undertaking.

Data Sources Appendix

Sources of data for Australian industrial disputes data are: Australian Bureau of Statistics (ABS) Industrial Disputes, Cat. No. 6321.0 various issues.

Sources of data for Indian industrial disputes data are: Labour Bureau, Government of India, Pocket Book of Labour Statistics Year 2003 & 2004, http://labourbureau.nic.in/pblsmp.html and Labour Statistics, http://labourbureau.nic.in/schemes.html and Datt (2003a)

Sources of labour force data: World Bank World Tables various issues available through EconData Pty Ltd.

Endnotes

(1.) The 'formal' or 'organised' sector is composed of the government sector plus private sector firms that employ more than 10 workers if using electric power or more than 20 if not using electric power. Private formal-sector firms are required to be registered with government. See Zagha (1999)

(2.) The distinguished economist Lord Desai has argued that the most effective, perhaps only, way to get legislative change happening in India is for India's two dominant political parties, the Congress Party and the BJP, to form a coalition to drive through necessary legislative change (South Asian Monitor 2005). Don't hold your breath.

(3.) The measure 'time lost' is generally considered to be not only the broadest measure of industrial disputes, but also the least likely to be affected by the problems of under-enumeration. See Turner (1969) for further discussion.

(4.) The use of the labour force as a divisor does not affect the overall shape of the series charted in the figures. It only affects the scale of the data. Note that the relative size of the formal versus informal sectors can vary according to different definitions that may be employed. For further discussion see Zagha (1999) and Tendulkar (2003) who defines organised labour as 'contractual, regular and hired employment' which he calculates to be about 14 per cent of the labour force.

(5.) If Indian strikes are applied (i.e. disputes minus lockouts) rather than industrial disputes, the correlation coefficient is 0.58.

(6) Profits, inventories and GDP growth have also been found to be significant determinants in Morris and Wilson (1999). Chapman and Gruen (1991) find the vacancy rate to be significant. Chapman and Gruen (1991), Morris and Wilson (1999) and Hodgkinson and Perera (2004) employ quite a number of one-off dummy variables to eliminate various perceived outliers. Since trends in these various determinants do not seem to explain the recent two decade decline in time lost due to disputes, which is the primary area of interest in this paper, we will not further pursue these determinants here.

(7.) See Sen Gupta and Sett (2000), Bhattacherjee (2001, 2002), Sodhi and Plowman (2001), Venkata Ratnam (2001) and Datt (2003a, 2003b)

(8.) Bhattacherjee (2001) drawing on an earlier study by Sen Gupta (1992) notes that '... as labour markets got tighter, ... the number of man days lost due to ... disputes increased phenomenally between 1966 and 1974 ...'

(9.) The strikers were led by Dr Datta Samant. According to Datt (2003b): 'This prolonged and irresponsible strike to satisfy the ego of a labour leader did tremendous damage to the working class ...' as when it eventually fizzled out it 'emboldened the employers [which resulted in a] ... rise in employer militancy to punish the working class' (p. 229). Samant (a qualified physician) was assassinated in 1997 by four gunmen. The strike itself is described by Bidwai (1997) as '... probably the greatest strike in the world, involving 200,000 workers and lasting 20 months. It happened when the workers' bargaining power was low and employers were looking for ways to close down mills'. As it turned out, the strike ended in a crushing defeat for the workers.

(10.) It must be noted that Indian trade union statistics are not considered to be particularly reliable. Venkata Ratnam (2001, p. 32) warns that: '... labour statistics in India are an eye wash. It is the footnotes, if any, that reveal more. The incidence of non-reporting of data or non-submission of returns is scandalously high even with respect to statutory returns'. In addition official data on union membership are typically published with a four year delay.

(11.) Note that different legislative/administrative eras have been identified by different researchers. Chapman and Gruen (1991) argue the Accord moderated industrial disputes. Morris and Wilson (1999) argue the Accord produced a moderation in disputes beyond the end-date of the Accord. And Hodgkinson and Perera (2004) argue that legislative changes of the 1990s produced a decline in industrial disputes.

(12.) For the USA, Japan, Germany, France, Italy, and the UK respective correlation coefficients with India for time lost for the period 1983-2002 are 0.6, 0.9, 0.6, 0.8, 0.8 and 0.8. For Australia they are 0.5, 0.8, 0.2, 0.5, 0.7 and 0.4 respectively. (Sources: International Labour Office (ILO), Yearbook of Labour Statistics plus ILO website: (http://laborsta.ilo.org/) and World Bank World Tables).

(13.) Indian employment and unemployment data need to be used with caution. The principal source of data on unemployment is the number of applicants currently registered at employment exchanges. See Asia Development Bank, 2004.

(14.) India experience of unemployment, as reflected in the number of applicants registered at employment exchanges divided by World Bank World Tables' estimates of the labour force, can be divided roughly into three distinct phases. The first phase is the 1960s. During this period the unemployment rate was relatively low and stable averaging 1.3 per cent The second phase was one of steadily rising unemployment. The rate rose from 2 per cent in 1971 to 10 per cent in 1991, the year of crisis and reform. The unemployment rate averaged nearly 6 per cent during this phase. The third phase is from 1992 to 2003. During this phase the unemployment rate averaged 9.2 per cent, though the rate did decline a little during this phase.

(15.) Australia's inflation trended down much more sharply than India's after the early 1970s. The high correlation between disputes in Australia and inflation does not apply to India. The respective correlation coefficients for frequency and inflation are .54 and .19. The respective correlation coefficients for time lost and inflation are .73 and .30.

(16.) For further discussion of Australian dispute duration see Sodhi and Plowman (2001), Dabscheck (1995) and Oxnam (1975, p.30) who observed that: 'No longer are strikes being employed mainly as trials of strength between disputing parties; instead they are being employed as media for ventilating protests against management policies and practices, and decisions of governments and their agencies considered detrimental to labour ....' Arguably, not a great deal has changed since Oxnam wrote.

(17.) The correlation coefficient for inflation and time lost due to strikes is 0.4, when 1982 (the year of the Bombay Textile strike) is excluded. The corresponding correlation coefficient for lockouts is 0.06

(18.) Datt's chosen periods of policy difference are (i) 1961 to 1975: the pre-emergency rule period, (ii) 1975 to 1990: the post-emergency rule period and (iii) 1991-2000: the New Economic Reform Period.

(19.) For a critical review of the use to which Perry (2005) puts Table 2 data, see Briggs (2005).

References

Asian Development Bank (2004), Key Indicators 2004, Asia Development Bank, Manilla

Australian Bureau of Statistics (2004a), Employee Earnings, Benefits and Trade Union Membership Australia, Cat No. 6310.0.

Australian Bureau of Statistics (2004b), Labour Force Australia, Cat. No. 6202.0

Australian Bureau of Statistics (2004c), Industrial Disputes, Cat. No. 6321.0

Bhagwati, J. (1993), India in Transition: Freeing the Economy, Oxford, Clarendon Press.

Bhattacherjee, D. (2001), 'The Evolution of Indian Industrial Relations: A Comparative Perspective', Industrial Relations Journal, vol. 32, pp. 244-263.

Bhattacherjee, D. (2002), 'Organised Labour and Economic Liberalisation in India: Past, Present and Future', in Jose, A. V. (Ed.), Organised Labour in the Twenty First Century, International Institute of Labour Studies, Geneva, pp. 307-345.

Bidwai, P. (1997), 'More than Murder', reproduced from Mid-Day, Mumbai, 27 January at http://www.proxsa.org/economy/labor/orglabr.html. Accessed 17.07.2005.

Beasley, T. and Burgess, R. (2002), 'Can Labor Market Regulation Hinder Economic Performance? Evidence form India', CERP Discussion Paper No. 3260.

Briggs, C. (2004a), 'The Return of the Lockout in Australia: a Profile of Lockouts since the Decentralisation of Bargaining', Australian Bulletin of Labour, vol. 30, pp. 101-112.

Briggs, C. (2004b), 'Lockout Law in Australia: Into the Mainstream', ACIRRT working paper 95, University of Sydney.

Briggs, C. (2004c), 'The Return of Lockouts Down Under in Comparative Perspective: Globalisation, the State and Employer Militancy', http://www2.umist.ac.uk/management/ewerc/wes2004/papers/briggs.pdf.

Briggs, C. (2005), 'Why Lockouts Matter: A Response to Perry', Australian Bulletin of Labour, vol. 31, pp. 40-43.

Chapman, B. J. and F. Gruen (1991), 'An Analysis of the Australian Consensual Incomes Policy: the Prices and Incomes Accord', in de Neubourg, C. (ed.), The Art of Full Employment (Elsevier, Amsterdam).

CIA (2004), The World Factbook, US Cental Intelligence Agency, http://www.cia.gov/cia/publications/factbook/

Davis, B. (2000), 'India and the East Asia Miracle', Agenda, vol. 7, pp 67-76.

Department of Foreign Affairs and Trade (DFAT) (2002), Country Fact Sheets for India and Australia, www.dfat.gov.au, accessed 15.04.2003

DFAT (2004), Country Fact Sheets for India and Australia, www.dfat.gov.au, accessed 01.02.2005

Dabscheck, B. (1995), The Struggle for Australian Industrial Relations, Oxford University Press, Melbourne.

Datt, R. (1993), 'Lockouts, Closures and the Role of the State', in Papola, T. S., Ghosh, P. P. and Sharma, A. N. (eds.), Labour Employment and Industrial Relations in India, B. R. Publishing Corporation, New Delhi.

Datt, R. (2000), 'Disturbing Trends in Strikes, Lockouts and Closures' Labour File: A Monthly Journal Of Labour And Economic Affairs, Note: Dart is misspelt as Dutt, http://www.labourfile.org/cec1//labourfile/lf_2000/Article2.htm

Datt, R. (2003a), Lockouts in India, New Delhi, Manohar.

Datt, R. (2003b), Economic Reforms, Labour and Employment, Delhi, Deep & Deep Publications.

International Labour Office (2003), Statistics of Trade Union Membership, Unpublished, ILO Statistics Bureau.

Healy, J. (2002), 'Peace at Last: Recent Trends in Australia's Industrial Action', The Australian Bulletin of Labour, vol. 28, pp 80-87.

Healy, J. (2004), 'The Australian Strike Rate and Industrial Relations: A Brief Reply to Perry', The Australian Bulletin of Labour, vol. 30, pp 149-54.

Hodgkinson, A. and Perera, N. (2004), 'Strike Activity Under Enterprise Bargaining: Economics or Politics', Australian Journal of Labour Economics, vol. 7, pp. 439-457.

Krueger, A. O. and Chinoy, S. (2002), 'The Indian Economy in Global Context', in Krueger, A. O. (ed.), Economic Policy Reforms and the Indian Economy, University of Chicago Press, Chicago, pp. 9-46.

Kuruvilla, S., Das, S., Kwon, H. and Kwon, S. (2002), 'Trade Union Growth and Decline in Asia', British Journal of Industrial Relations, vol. 40, pp. 431-461

Labour Bureau Government of India (2004a), Pocket Book of Labour Statistics Year 2003 & 2004, http://labourbureau.nic.in/pblsmp.html

Labour Bureau Government of India (2004b), Labour Statistics, http://labourbureau.nic.in/schemes.html

Mathur, A. N. (1991), Industrial Restructuring and Union Power, ILO-ARTEP New Delhi.

Morris, A. and Wilson, K. (1999), 'Industrial Disputes and the Accord: A Final Word', Australian Bulletin of Labour, vol. 25, pp. 63-71.

Oxnam, D. W. (1975), 'The Incidence of Industrial Disputes in Australia', in Isaac, J. E. and Ford, G. W. (eds), Australian Industrial Relations: Readings, 2nd Edition, Sun Books, Melbourne.

Perry, L. J. (2004), 'Comments on the Decline in Strikes in Australia', Australian Bulletin of Labour, vol. 30, pp. 134-148.

Perry, L. J. (2005), 'The Return of the Lockout', Australian Bulletin of Labour, vol. 31, pp. 33-39.

Perry, L. J. and Wilson, Patrick J. (2001), 'The Accord and Industrial Disputes: An International Perspective', Australian Journal of Labour Economics, vol. 4, pp. 232-247.

Perry, L. J. and Wilson, Patrick J. (2003) 'An Analysis of International Linkages in Strike Activity', International Journal of Employment Studies, vol. 11, pp. 47-74.

Perry, L. J. and Wilson, Patrick J. (2004), 'Trends in Work Stoppages: A Global Perspective', Working Paper No. 47, November, Policy Integration Department, Statistical Development and Analysis Unit, International Labour Office. http:// www.ilo.org/public/English/bureau/integration/departme/about/working.htm

Rees, A. (1952), 'Industrial Conflict and Business Fluctuations', Journal of Political Economy, vol 60, pp. 371-382.

Sen Gupta, A. K. (1992), Trends in Industrial Conflict in India (1961-87), Friedrich Ebert Stiftung, New Delhi.

Sen Gupta, A. K. and Sett, P. K. (2000), 'Industrial Relations Law, Employment Security and Collective Bargaining in India: Myths, Realities and Hope', Industrial Relations Journal, vol. 31, pp. 144-153.

Sanyal, P. and Menon, N. (2004), 'Labor Regulation and the Economics of Firm Geography: Evidence From India', Working Paper available from the SSRN Electronic Paper Collection.

Sharma, A. N. (2004), 'Globalisation, Labour Markets and Poverty: Towards an Agenda for Decent Work in India', paper prepared for the 1st Global Labour held during the 5th Asian Regional Congress of the International Industrial Relations Association, June 23-26.

Sodhi, J. S. and Plowman, D. H. (2001), Economic Change and Industrial Relations: India and Australia, Scholastic Press Australia, Nedlands.

South Asian Monitor(2005), 'Amartya Sen, Meghnad Desai Debate Idea of India' 4 January, http://www.southasiamonitor.org/india/2OO5/jan105ind1.shtml

The Economist (1997), 'A Survey of India', 22 February.

The Economist (2005), 'The Tiger In Front', 5 March.

Tendulkar, S. D. (2003), 'Organised Labour Markets in India Pre and Post Reform', Conference on Anti Poverty and Social Policy in India, MacArthur Research Network on Inequality and Economic Performance, Neerama Fort Palace, Alwar, Rajasthan, January 2-4.

Turner, H. A. (1969), Is Britain Really Strike Prone? A Review of the Incidence, Character and Costs of Industrial Conflict, Occasional Paper No 20, University of Cambridge, Department of Economics, Cambridge UP, Cambridge.

Venkata Ratnam, C. S. (2001), Globalisation and Labour-Management Relations Dynamics of Change, Response Books, New Delhi.

Walsh, K. (1983), Strikes in Europe and the United States: Measurement and Incidence, Frances Pinter, London.

World Bank (2004), World Bank World Tables, Washington.

Zagha, R. (1999), 'Labour and India's Economic Reforms' in Sachs, J. and Varshney, A. (eds), India in the Era of Economic Reforms, Oxford University Press, New Delhi.

L.J. Perry, School of Finance and Economics, University of Technology, Sydney

Table 1: India-Australia Comparisons

Descriptor                                       Period   India  Aust

Population: millions (a)                          2003    1062    20

Population Density: people per square             2003     319   2.5
  kilometre (a)
GDP: Purchasing power parity, % USA (b)           2002     27     5
Per Capita GDP: Purchasing power parity,          2002      7     79
  % USA (b)
Real GDP Growth: Local currency, annual          1961-03  4.7%   3.7%
  average (a)(c)
Real per capita GDP Growth: Local currency       1961-03  2.6%   2.1%
  (a)(c)
Real per capita GDP Growth: Local currency       1961-91  1.9%   2.0%
  (a)(c)
Real per capita GDP Growth: Local currency       1992-03  4.2%   2.6%
  (a)(c)
Inflation: Consumer price index, annual average  1961-03  7.8%   5.7%
  (a)(c)
Inflation: Consumer price index, annual average  1992-03  7.4%   2.4%
  (a)(c)
Union Density: % of employees, annual average    1980-00  23 *    39
  (d)(e)(f)
Union Density: % of labour force, annual         1980-00   1.6    29
  average (c)(d)(e)(f)
Union Power Index: Kuruvilla et al. (2002)       1980-95   15    402
  % USA (g)
Unemployment rate (a)                             2003    9.5%    6%
Agriculture Value Added (% GDP) (c)               2002     23     4
Agriculture Labour Force (% Total) (h)            2002     60     5
Urbanisation Rate (c)                             2002     28%   92%
Female Participation Rate (c)                     2002     33%   44%
Life Expectancy in years at Birth (c)             2002     63     79
Human Development Index World Rank (b)            2002     127    3
Human Development Index World Rank, % USA (b)     2002     63    101

(a) DFAT (2004), (b) HRDP (2004), (c) World Bank (2004),
(d) ILO (2003), (e) Labour Bureau Government of India (2004a),
(f) Australian Bureau of Statistics (2004a, 2004b),
(g) Kuruvilla et al. (2002), (h) CIA (2004).

* Formal sector employees, these are a fraction of total employees

Table 2: Working Days Lost Due to Strikes and
Lockouts in Australia (Thousands)

                 All Industries
           Strikes   Lockouts   Total
Period       (c)       (b)       (a)

1994-98    2989.9      48.6     3038.5
1999-03    2005.8     205.7     2211.5
% change     -33       323       -27

                 Manufacturing
           Strikes   Lockouts   Total
Period       (f)       (e)       (d)

1994-98     609.3      18.7     628.0
1999-03     536.4     194.5     730.9
% change     -12       940        16

               Non-Manufacturing
           Strikes   Lockouts   Total
Period       (g)       (h)       (i)

1994-98    2380.6      29.9     2410.5
1999-03    1469.4      11.2     1480.6
% change     -38       -63       -39