Worker turnover and job reallocation: the role of fixed-term contracts.
The increasing interest in the dynamics of labour markets since the 1970s has generated much work on the computation of worker and job turnover. Such work has concluded that the amount of job reallocation (the sum of job creation and destruction) is large in OECD countries. Moreover, new evidence from databases incorporating information on the number of hirings and separations at firm or establishment-level suggests that workers mostly enter jobs that existed before their appointment and that will continue to exist after they quit or are fired. In other words, the rotation component of worker flows accounts for a large proportion of total worker turnover.
Are those findings in accordance with homogeneous firm behaviour (the representative firm assumption in labour demand models)? The answer to this question requires the analysis of the likely relationship between net employment changes and hirings/separations (i.e. whether employment growth (decline) and hirings (separations) are interchangeable terms for the same phenomenon), and the simultaneity of hirings and separations at firm-level. Empirical work on these issues is limited. Only Hamermesh et al. (1994) and Lane et al. (1996) have provided some evidence. Their results suggest that those terms are not interchangeable and that there are simultaneous worker flows.
The contribution of this paper is to address all the above-mentioned issues for the Spanish case while taking into account the fact that employment relationships between firms and workers can be of two types: permanent contracts and fixed-term contracts. In most OECD countries, the use of the latter has increased substantially during the 1980s and 1990s (see Bentolila and Dolado, 1994).(1) The distinction between the type of contracts highlights the difference in separation costs. In Spain, firms need collective layoffs to be approved before-hand by the corresponding labour authority, and they pay severance payments to workers with permanent contracts if workers are laid off. For temporary workers, firms know exactly the amount to be paid as separation costs (close to zero), but those costs are subject to bargaining in the case of permanent workers.
The data used in this paper is a sample of large plants (more than 500 employees) for the period 1993-4 from a database, the Survey of Economic Situation (Encuesta de Coyuntura Laboral, ECL), launched by the Spanish Ministry of Labour and Social Affairs in 1990. This is the first time that dataset has been used. Its main advantage is that it collects information at the establishment level for the non-agricultural economy, providing quarterly information on the total number of workers arriving at and leaving establishments by type of contract. However, given the specific group of establishments included in the sample, the results should be interpreted with care as they cannot be extended to the whole Spanish economy.(2)
The structure of the paper is as follows. Section 2 provides the definitions used throughout the paper, surveys the relevant literature, and describes the data in more detail. The empirical analysis proceeds in the next two sections, where the role of fixed-term contracts is addressed in detail. In Section 3 the relationship between net employment changes and hirings/separations is examined using simple cross-tabulations of the data. Next, in Section 4 the extent of worker turnover and the relative importance of its components is documented. Section 5 concludes.
2. Definitions, previous literature, and data 2.1 Definitions
Total worker turnover or worker reallocation (WR) is defined as the formation and dissolution of employee-employer job matches, thus consisting of total hirings (H) and separations (S) in the economy: WR H + S. Worker reallocation can be divided into two components. The first element arises as a consequence of firms creating and destroying job positions (gross job reallocation or job turnover, JR), leading to changes in the level of employment. The second component of turnover occurs independently of job flows, with no effect on the level of firms' employment positions. This sort of worker flows is called rotation (RR). This is a result of workers moving between job positions and may be either worker-initiated (due to personal factors, low job satisfaction, higher promotion expectations in alternative firms) or firm-initiated (as firms replace workers without changing employment positions, perhaps because job matches are poor).
How are those measures of turnover estimated? And how are they related to net employment changes? First, the net employment change ([Delta]E) in an establishment is measured as the difference between total hirings and total separations. It may be decomposed using the flows of workers as follows
[Delta]E [equivalent to] H- S [equivalent to] H- (Q+L+ T+R) (1)
where total separations consist of quits (Q), layoffs (L), transfers of workers to other plants belonging to the same firm (T), and retirements (R).
Firms may use two types of contract when hiring employees: permanent and fixed-term contracts. With this distinction, the previous identity can be rewritten in the following way
[Delta]E [equivalent to] H - S [equivalent to] ([H.sub.p] + [H.sub.F] - (Q + L + C + T + R) (2)
where total hires are now comprised of permanent hires ([H.sub.P]) and temporary hires ([H.sub.F]). In addition, termination of fixed-term contracts (C) is a new source of separations.
Net employment change may also be computed using gross flows based on positions, not people. Thus, the net employment change in an establishment is the difference between the number of jobs in periods t and t + 1. It also may be defined as the difference between job creation and job destruction in that interval of time:
[Delta]E [equivalent to] [J.sub.t+1] - [J.sub.t] [equivalent to] JC - JD.
Moreover, the sum of job creation and job destruction gives the gross job reallocation or job turnover: JR = JC + JD. It measures the reallocation of employment positions across and within establishments. In fact, as our dataset does not contain information on positions, job reallocation is proxied by plant-level net employment changes (or disaggregate job flows). This measure has been used in recent empirical work as a measure of gross job flows (see, for instance, Davis and Haltiwanger, 1990). Gross job reallocation is calculated by summing employment gains at expanding plants and employment losses at shrinking plants:
JR[prime] = JPOS + JNEG = [Delta][E.sup.+] + [Delta][E.sup.-].
See the Appendix for a description of the methodology used to estimate these flows.(3)
Finally, it is possible to relate measures based on positions to measures based on people. The relationships are as follows
[Delta]E [less than or equal to] JPOS + JNEG [less than or equal to] JC + JD [less than or equal to] H + S (3)
2.2 Previous literature
The availability of firm or establishment-level databases has been the main factor giving rise to empirical studies on gross employment flows. Such studies include: Leonard (1987); Dunne et al. (1989); Blanchard and Diamond (1990); Davis and Haltiwanger (1990, 1992); and Davis et al. (1994) in the US; Contini et al. (1992) in Italy; Boeri and Cramer (1992) in Germany; Blanchflower and Burgess (1993) and Konings (1995) in the UK; and Dolado and Gomez (1995) in Spain.
This empirical work has found some stylised facts for different national labour markets (see Contini et al., 1995, for a comprehensive study), although results should be taken with care due to differences in sample periods, industries, and units of analysis. Firstly, the level of job creation and job destruction is very large, and there is simultaneous job creation and job destruction even within narrowly defined aggregates (either sectors or regions). Secondly, for some countries (the US and the UK) job creation moves cyclically and job destruction moves countercyclically; that leads to an asymmetry of gross job flows in the business cycle. However, for another group of countries (Italy, Germany, and Spain) job creation equals job destruction across the cycle, i.e. there are symmetrical gross job flows over the business cycle. Thirdly, as a result of the previous issue, the gross job reallocation gives varying results for different countries. In the case of the US and the UK, the job reallocation rate moves countercyclically: it increases in recessions and decreases in booms. However, for the other group of countries job reallocation rates do not respond to the economic conditions. These differing results may be related to the flexibility degree of labour markets: it seems that the labour markets of continental countries behave differently than Anglo-Saxon countries. Finally, these studies have shown that there are some firm-specific characteristics (mainly age, size, and industry) affecting the levels of job reallocation.
One obvious drawback of the databases in these studies is the lack of information on the number of hirings and separations for each firm. This limitation means it is impossible to calculate total worker turnover and, therefore, to relate the amount of worker mobility due to job reallocation to total worker turnover. The examination of new databases incorporating that information has generated new studies: Anderson and Meyer (1994) for the US; Hamermesh et al. (1994) for the Netherlands; Lagarde et al. (1994) for France; and OECD (1996) summarising the results for OECD countries. Their findings suggests that workers mostly enter jobs that existed before and that will exist after they leave. In other words, the rotation component of worker flows accounts for the largest portion (60-70%) of total worker turnover.
Moreover, Hamermesh et al. (1994) and Lane et al. (1996) have analysed the relationship between net employment changes and hiring/firing status at firm level to study whether these terms are interchangeable and their likely simultaneity. They use annual data for the Netherlands in 1988-90 and quarterly data for the state of Maryland in 1985-93, respectively. Their results suggest that hiring is not restricted to expanding firms, separations (particularly, firings) are not restricted to shrinking firms, most firing is done by firms which are also hiring, and the large majority of worker mobility is to and from existing employment positions. In conclusion, the above-mentioned terms are not interchangeable and there are simultaneous worker and job flows.
Previous empirical work has not considered, however, the role of temporary contracts in bringing about worker and job turnover, mainly due to the lack of information on the employment relationship between firms and their workers. The dataset used in this paper does contain this sort of information, as temporary work is a distinctive feature of the Spanish labour market. Therefore, the consideration of contractual status is the main contribution of this paper in the literature.
The data source in this paper is a survey of firms (Survey of Economic Situation or Encuesta de Coyuntura Laboral, ECL). It has been carried out quarterly by the Spanish Ministry of Labour and Social Affairs since the second quarter of 1990 and the unit of analysis is the establishment with more than five employees. The data refer mainly to single-establishment firms and multiple-establishment firms that report separately information for the subordinate units, but it is possible that some multiple-establishment firms combine all information on subordinate units into a single entity. The database covers non-agriculture industries but excludes Public Administration, Defense and Social Security, diplomatic delegations, and international and religious organizations from the service sector.
In particular, this paper uses information on establishments having more than 500 employees (the number of such large plants being less than 1,000) who have answered the questionnaire in all of the quarters of the period 1993-4. This is the only group of plant size and the only period for which data have been provided by the Spanish Ministry of Labour and Social Affairs. The number of units contained in the sample (panel) is 737, representing around 15% of total non-agriculture employment and having a proportion of fixed-term workers of 20%.(4)
The main characteristics of the dataset concerning the calculation of worker and job flows are described below. At the end of each quarter establishments provide information on the employment stock at that moment and at the end of the previous quarter. They also provide data on the number of arrivals and separations by type of contract during that period, somewhat novel when compared to other similar international databases.
As for the information on the number of arrivals and separations, it is worth noting that more than one move for the same worker during the same quarter may be included, but not those moves occurring within the same month of a given quarter. For instance, if a worker is hired in the first month of a given quarter, he/she leaves in the second month, and he/she is rehired in the third month, the employer counts two arrivals and one separation. But if a worker is hired, separated out, and rehired within a month, the employer only counts one arrival following the survey guidelines. This means that the ECL does not give information on the shortest worker flows, affecting those with temporary contracts lasting only a few days or weeks. Therefore, the survey may be underestimating worker turnover for short transitions in a given month.
On the other hand, there is the issue of the extent of rehiring and renewal of contracts. If a firm renews a fixed-term contract when it is about to end, it is assumed that the employer does not count one separation and one hire, because the worker has not left in order to be rehired; thus, the database would consider that case as continuous employment. Therefore, although it is possible that firms renew temporary contracts of some workers when they are about to end (the Spanish law allows it with some restrictions),(5) the database does not collect those cases. However, if an intervening, short period of unemployment exists between the termination of a fixed-term contract and a rehire, the latter would be captured by the database. One may say that rehires are not worker turnover because it simply counts one separation and one hire that are not 'real' (in the sense that the same worker keeps on working for the same firm). From this point of view, worker turnover would be overestimated if these results were included.
Employers also give information on the different type of separations: layoffs (consisting of dismissals and redundancies), quits, end of contracts, legal and early retirements, and transfers to other plants within the same firm. Distinction among these types of job separation are the result of employers completing the survey questionnaire and, therefore, subject to their interpretation. This raises many problems. The most obvious is that conceptually it is very difficult to tell exactly whether a separation is a quit (voluntary for the workers and involuntary for the firm) or a layoff (involuntary for the workers and voluntary for the firm). It is possible as well that a layoff may be hidden either as an incentive quit or as an early retirement. Similar arguments may be made for termination of temporary jobs (see Pissarides, 1986, on all these issues).
Finally, the database does not contain information on newly created and exiting firms. Therefore, it is impossible to analyse the contribution of firm births and deaths to worker and job reallocation. Although this limitation may be important in order to fully characterize worker turnover, it must be stressed that other databases suffer the same problem (OECD, 1994). Moreover, the effect of not having information on newly created and exiting firms is not very important when considering large firms - as in this paper - since most births and deaths are concentrated in smalls firms (Contini et al., 1995).
3. Net employment changes, gross worker flows, and the simultaneity of hirings and separations
In the next two sections, I use the pooled sample of 5,896 observations for large establishments to deal with the empirical analysis of two main issues. In Section 4, the focus is on the relationship between net employment changes and the flow of workers in and out of firms, and the analysis of the simultaneity of hirings and separations. In Section 5, I examine the components of worker turnover, i.e. turnover due to job reallocation and turnover in excess of job reallocation (rotation). In both sections, the role of fixed-term contracts is considered in detail.
Table 1 provides estimates of the flows in eq. (1). It gives the average quarterly hiring and separation rates (and the corresponding layoff and quit rates) for the total sample and for three groups of establishments: those with growing, stable, and declining employment. While the overall hiring rate is 6.6%, the separation rate is 7.1%.
There are three main findings emerging from information in Table 1. Firstly, the hiring rate is substantially higher among expanding establishments than among declining establishments. However, while growing firms have a hiring rate twice as large as the average, the corresponding rate for declining firms is still 4%. In fact, growing establishments account for only 63% of all hirings.
Secondly, somewhat surprisingly the separation rates hardly differ between both groups. This means that 7-8% of the workforce in declining and expanding firms [TABULAR DATA FOR TABLE 1 OMITTED] leave each quarter. Moreover, declining establishments account for only 40% of all separations.
Thirdly, quits are higher in growing establishments than in declining establishments. The opposite is true in the case of layoffs. However, the basic finding here is that layoffs and quits only account for less than one tenth of all separations. How is that possible? The answer must be sought in the use of fixed-term contracts. For this reason, Table 2 replicates Table 1 but divides the hiring and separation rates into two parts: that corresponding to permanent contracts and that corresponding to fixed-term contracts. In addition, separations other than layoffs and quits have been considered. Therefore, Table 2 presents estimates of the flows in eq. (2).
As expected, total hiring and separation rates are mainly related to the use of fixed-term contracts (85% and 79%, respectively). The previous findings related to [TABULAR DATA FOR TABLE 3 OMITTED] hiring rates still hold, i.e. the hiring rate is higher among expanding establishments than among declining establishments. This is true for either permanent or fixed-term contracts. With respect to separation rates, a composition effect is observed: while the separation rate of fixed-term contracts is higher among growing establishments than among declining establishments, the opposite is true in the case of the separation rate of permanent contracts.
Table 3 further investigates the relationship between employment growth, hirings, and separations. This table groups establishments according to net employment status (growing, stable, and declining employment) and to hiring and separation status (four possible combinations). Given the importance of distinguishing between permanent and fixed-term contracts, two different type of separations have been considered: layoffs (affecting the former) and termination of contracts (affecting the latter). As can be seen, employment declined in nearly two thirds of establishments in an average quarter. The upper panel shows that more than 90% of the establishments were hiring in a given quarter; but while 77% were simultaneously separating workers (either through layoffs or through termination of contracts), only 15% were not simultaneously separating workers. In addition, only a small fraction (3.6%) of total establishments were separating workers without hiring.
[TABULAR DATA FOR TABLE 4 OMITTED]
Not surprisingly, the previous findings are driven by the fact that most hirings and separations are related to fixed-term contracts, as can be seen by observing the similarity of those figures in the bottom panel and those in the upper panel. However, when only layoffs are considered as separations, the results are different. As before, more than 90% of the establishments were hiring, but now the distribution has changed: more than one third (37.6%) laid off simultaneously and more than one half (54%) did not use layoffs. And only a small fraction (1.8%) of total establishments were laying off without hiring in a quarter.
Given the previous results, Table 4 analyses to what extent hirings and separations of workers with permanent and fixed-term contracts are simultaneous, and whether those hirings and separations can be attributed to the existence of that sort of segmentation in the labour market. For that reason, establishments have been grouped in four hiring regimes and in four separation regimes. First of all, 70% of all plants are hiring permanent workers, of interest given the extensive use of fixed-term contracts in Spain. Second, consistent with Table 3, 77% of all plants (those figures in bold) are simultaneously hiring and separating. Among those, 25.6% are simultaneously hiring and separating of workers using both type of contracts, and 36% (12.8% + 23.1%) are hiring and separating temporary workers. Among the 77% of establishments simultaneously engaged in hiring and separating, only 3.9% (3.3% + 0.6%) of all establishments are hiring using one type of contract and separating using the other. Thus, it seems that the alleged segmentation in the labour market only accounts for a small portion of simultaneous hires and separations at plant-level data, as far as large firms are concerned.
In order to investigate further whether firms substitute temporary workers for permanent workers, it is necessary to establish some definitions. Net permanent employment growth is defined as the difference between the number of permanent hirings and the number of layoffs: [Delta][E.sub.P] = [H.sub.P] - L. Similarly, net temporary employment growth is defined as the difference between the number of temporary hirings and the number of termination of contracts: [Delta][E.sub.P] = [H.sub.F] - C. Finally, I consider that substitution is at work when a plant is simultaneously decreasing its level of permanent employment and increasing its level of temporary employment in a given quarter. The result is that, among those establishments that are simultaneously hiring and separating, only 7% are substituting temporary workers for permanent workers in a given quarter. Therefore, the substitution at the aggregate (temporary employment slightly increased while permanent employment declined for the sample of large plants) is not widely observed with plant-level data. This seems to indicate that firm behaviour is heterogeneous and that interplant substitution has been more important than intraplant substitution. One possibility is that interplant substitution has been between large and small plants. In order to corroborate this presumption, Table 5 shows the distribution of temporary contracts by establishment size over the period 1990-96 (information is provided for three groups: 6-50 employees, 51-250 employees, and more than 250 employees, as it is published by the Ministry of Labour and Social Affairs). Data show that during the sharp crisis of 1992-3 the share of large plants increased because temporary employment declined more intensely in small and medium-size plants than in large plants. Moreover, during the subsequent recovery the share of large plants kept on increasing; this was due to small plants increasing permanent employment and reducing temporary employment and large plants doing the opposite. Therefore, it seems that employment adjustment behaviour differs by establishment size.
Table 5 Distribution of fixed-term contracts by establishment size period 6-50 51-250 [greater than]250 total II-90/II-92 56.3 25.4 18.3 100.0 II-92/II-94 6.6 4.3 19.1 100.0 II-94/IV-96 54.5 25.1 20.4 100.0 Data source: ECL (1990-6)
4. The components of worker turnover: job reallocation and rotation
The basic finding in the previous section is that in a given quarter workers are simultaneously arriving at and leaving most of the establishments. Now, the paper focuses on a related issue. Is worker mobility mainly due to job reallocation of employment positions? Or, alternatively, does worker mobility reflect a continuous sorting and re-sorting process of workers across a given set of jobs? As before, the role of fixed-term contracts is also addressed throughout this section.
Given the fact that the dataset does not contain information on creation and destruction of employment positions, a standard methodology has been applied in order to estimate and separate out the portion of worker turnover that is due to job reallocation and rotation. The Appendix presents a formal definition of job [TABULAR DATA FOR TABLE 6 OMITTED] creation, job destruction, and job reallocation concepts. Job creation (JPOS) is defined as the sum of employment gains at growing establishments, and job destruction (JNEG) as the sum of employment losses at declining establishments. Gross job reallocation (JR) is the sum of job creation and job destruction, while net job creation (NET) is the difference.
Table 6 provides the basic results and estimates the flows in eq. (3). Columns (3) and (4) are familiar to us. In fact, WPOS is the hiring rate and WNEG is the separation rate I estimated in Table 1. The sum of both gives total worker turnover (WR). Thus, in an average quarter nearly 14% of workers arrive at or leave establishments. Moreover, average job creation (column 1) is only 1.3% of total employment, while average job destruction (column 2) is larger (1.9%). Thus, gross job reallocation (column 5) is 3.3% of total employment.
Worker turnover arising as a consequence of reallocation of employment positions may be related to total worker turnover: the former only accounts for one fourth of the latter. This means that the majority (75%) of observed worker reallocation reflects external rotations on positions that are neither created nor destroyed. This finding is similar to those obtained by other empirical work for OECD countries (see Table 7). As can be seen, gross job reallocation in OECD countries represents 25-40% of total worker turnover.
Table 6 also provides information on job and worker flows by type of contract. Given the findings in the previous section, that distinction makes sense, as it is expected that gross flows related to fixed-term contracts should be higher than those corresponding to permanent contracts. Some basic findings are worth noting.
First, job reallocation is only 1.6% of total permanent employment as compared with 11.2% of total temporary employment. However, job destruction accounts for a larger portion of job reallocation in the case of permanent jobs when compared with temporary jobs: 74% against 48%. The net result is a decline of permanent employment and an increase of temporary employment.
Second, total worker turnover is 2.6% for permanent employment and 56.7% for temporary employment. This means that in an average quarter a very small portion of permanent workers move, but more than half of all temporary workers arrive at [TABULAR DATA FOR TABLE 7 OMITTED] or leave establishments. This gives an idea of the short duration of many employment matches and the segmentation at work in the Spanish labour market. The result to be stressed here is that while job reallocation accounts for 60% of total worker turnover for permanent workers, that proportion is only 20% for temporary workers. Therefore, permanent contracts are mainly linked to creation and destruction of employment positions (although 40% of total worker reallocation is related to rotation among existing jobs), and fixed-term contracts are mainly used for rotation purposes, i.e. to vacate and refill existing posts.
Finally, as a summary of previous information, the bottom row of Table 6 presents the ratio of temporary contracts' flows to permanent contracts' flows. As can be seen, worker turnover for temporary workers is 23 times the turnover for permanent workers. This is due to the enormous number of rotation flows for temporary workers. In contrast, job reallocation for temporary workers is only eight times that for permanent workers. This is due to job destruction linked to permanent positions. In the aggregate, however, fixed-term contracts also account for the majority of job creation and destruction.
5. Concluding remarks
This paper has addressed for the first time in the literature the role of fixed-term contracts in bringing about simultaneous worker flows and worker and job turnover. The analysis has been carried out for a specific group of establishments in the Spanish labour market (namely, those plants with more than 500 employees) using a database to compute gross flows of positions and gross flows of workers by type of contract. Some empirical issues related to the dynamics of labour markets with implications for economic theory and policy have been examined.
The main results are as follows. First, net employment growth (decline) and hirings (separations) are not interchangeable terms, as it has been detected that hirings (separations) are not restricted to expanding (shrinking) establishments. Moreover, when looking at the contractual status it is observed that fixed-term contracts account for the largest portion of the hiring and separation rates.
Second, most establishments are simultaneously hiring and separating workers. The existence and extensive use of temporary contracts allows firms to adjust to shocks, although in some cases these are so large that they must rely on layoffs. Firm heterogeneous behaviour is detected in that an important proportion (26%) of all establishments are simultaneously hiring and separating workers using both type of contracts but only 4% of all are hiring using one type of contract and separating with the other. In addition, the observed substitution at the aggregate level (temporary employment slightly increased while permanent employment declined for large establishments in the period of analysis) is not fully detected with plant-level data. It seems that interplant substitution (between large and small establishments) has been more important than intraplant substitution.
Third, although job reallocation and worker turnover rates are both higher for temporary employment, the impact of the former differs markedly by type of contract: permanent contracts are mainly linked to creation and destruction of employment positions and fixed-term contracts are mainly used for rotation purposes. However, at the aggregate level fixed-term contracts also account for the majority of job creation and destruction. These findings may be related to the existence of alleged large firing costs for permanent employees and various type of temporary contracts in the Spanish labour market. This combination may result in an increased incidence of short-term contracts and, thus, rotation (Booth, 1996).
All previous results are important as they have implications for dynamic models of labour demand, macroeconomic adjustment, and policy implications. Labour demand models (see Hamermesh, 1993) equate net employment growth with hiring and net employment decline with separations, ignoring the fact that many firms are simultaneously hiring and separating workers. From a macroeconomic point of view, the distinction between hiring/separations, job creation/destruction, and permanent/temporary employment is essential in the presence of heterogeneous labour and employment positions. Therefore, ignoring gross worker and job flows avoids taking into account the potential adjustment costs generated by demand shocks, since the costs to a firm are indicated not only by the net employment change but also by the costs of hiring and separating of permanent and temporary workers and the costs of creating and destroying positions (see Hamermesh, 1995, on these issues).
Finally, from a policy perspective, governments may try to increase job turnover (through relaxing employment protection legislation) and/or worker turnover (increasing the inflow into and the outflow from unemployment) to improve the situation for the long-term unemployed. But the level of employment adjustment is considerable across all countries and is not associated with cross-country differences in net employment growth (OECD, 1996). Therefore, it is not clear that either increased number of vacancies or increased rotation (using fixed-term contracts) will necessarily reduce long-term unemployment. And there is the issue of the effects of increasing labour turnover on the economy, namely on wage formation and labour force training and productivity. If temporary workers are not attached to the firm, they are less likely to receive training than permanent workers, and this can have a negative impact on labour productivity. In any case, further research and additional and better data sources are needed to understand intercountry differences of labour markets adjustment.
This research was carried out while the author visited the Institute for Employment Research (University of Warwick) and the ESRC Research Centre on Micro-social Change (University of Essex). Economic support from the Spanish Ministry of Education (FPI Programme, grant number EX95 03092396) and from the European Commission (HCM Programme, award number 04505) is gratefully acknowledged. The author is indebted to Miguel A. Malo for helpful discussions. He also thanks two anonymous referees, Alison Booth, and participants in seminars at FEDEA (Madrid, Spain) and Essex for their comments. Thanks are extended to the Spanish Ministry of Labour and Social Affairs, and particularly Maria Teresa Giraldez for their permission to use the dataset and Maria Angeles Perez Corrales for their helpful comments.
Here methodology used in the paper to analyse worker and job flows is explained. This methodology follows the empirical approach outlined in Davis and Haltiwanger (1990) which has been extensively used in most of the subsequent literature. Worker turnover (gross flows of workers) can be divided into two components: worker mobility related to gross job reallocation (due to job creation and job destruction processes), and worker mobility in excess of job reallocation (occurring independently of job flows). As our dataset gives information on external gross flows of workers (hirings and separations) at establishment level, it is possible to measure total worker reallocation and its components.
First of all, let us consider total worker turnover. Given the size of establishment i at times t ([E.sub.i,t]) and t - 1 ([E.sub.i,t-1]), the average size of establishment i between t - 1 and t is defined as follows: [N.sub.i,t] = ([E.sub.i,t] + [E.sub.i,t-1])/2. By aggregating, it is possible to obtain the size of the whole economy: [N.sub.t] = [[Sigma].sub.i][N.sub.i,t].
The hiring (separation) rate is defined as the proportion of the number of workers arriving in (leaving) establishments between t - 1 and t with respect to the employment stock: [h.sub.i,t] = [H.sub.i,t]/[N.sub.i,t]([s.sub.i,t] = [S.sub.i,t]/[N.sub.i,t]).
Then, by aggregating, we may calculate the aggregate hiring rate
[WPOS.sub.t] = [[Sigma].sub.i]([N.sub.i,t]/[N.sub.t])[h.sub.i,t]
and the aggregate separation rate
[WNEG.sub.t] = [[Sigma].sub.i]([N.sub.i,t]/[N.sub.t])[s.sub.i,t]
The sum of both rates is the turnover or worker reallocation rate ([WR.sub.t]). It gives us an idea of gross external mobility of workers in the whole economy
[WR.sub.t] = [WPOS.sub.t] + [WNEG.sub.t]
Now let us consider the case of job flows. First, the employment growth rate in each establishment is defined as follows: [g.sub.i,t] = ([E.sub.i,t] - [E.sub.i,t - 1])/[N.sub.i,t] = ([H.sub.i,t] - [S.sub.i,t])/[N.sub.i,t].. In other words, the employment growth rate can be calculated either as the difference between current and past employment stocks or as the difference between hirings and separations in the corresponding period of time. In both cases, it tells us the net job creation or destruction rate in each establishment.
Then, by aggregating, we define the job creation rate
[JPOS.sub.t] = [[Sigma].sub.i]([N.sub.i,t]/[N.sub.t])[g.sub.i,t], for [g.sub.i,t] [greater than] 0
and the job destruction rate
[JNEG.sub.t] = [[Sigma].sup.i]([N.sub.i,t]/[N.sub.t]) [multiplied by] [g.sub.i,t], for [g.sub.i,t] [less than] 0
The job reallocation rate ([JR.sub.t]) is the sum of both rates. It gives us an idea of the external mobility of workers (in the whole economy) which is due to job creation and job destruction
[JR.sub.t] = [JPOS.sub.t] + [JNEG.sub.t]
Now the rotation component of worker turnover across establishments ([RR.sub.t]) is computed. Since it is defined as the worker reallocation in excess of job flows, the following equation holds:
[RR.sub.t] = [WR.sub.t] - [JR.sub.t] = 2 min ([H.sub.t], [S.sub.t])
Finally, aggregate net employment growth rates will be the difference between both job creation and job destruction. It also may be computed as the difference between total hirings and total separations
[NET.sub.t] = [JPOS.sub.t] - [JNEG.sub.t] = [WPOS.sub.t] - [WNEG.sub.t]
1 In Spain, fixed-term contracts have been increasingly used by firms since a change in the labour legislation took place at the end of 1984 to increase labour market flexibility. The change in the law permitted temporary contracts to be used for reasons other than the temporary nature of the job. The aim was that firms could adjust employment more easily without having to change separation rules for permanent contracts. The result was that the proportion of fixed-term workers increased very quickly until the early 1990s, when one in three employees were working on a temporary basis. Since then the proportion has tended to remain more or less stable.
2 The state of the Spanish labour market in the period of analysis may also affect the results. During 1992-3, the Spanish economy suffered a short but sharp crisis, reaching an unemployment rate record figure of 24% at the beginning of 1994 and remaining nearly stable during that year (Labour Force Survey figures).
3 This way of calculating job reallocation implies that JR' is a lower bound of real gross job flows (JR). The reason is that, as creation and destruction of positions cannot be observed, the researcher considers that an establishment has created (destroyed) jobs if the difference between current and past employment stocks is positive (negative). That excludes job reallocation within firms. For instance, restructuring that causes simultaneous job creation and job destruction, but leaves employment constant, will be missed. Moreover, the measures of job creation and job reallocation do not include unfilled vacancies created by new positions. However, as a counterbalance, they include filled vacancies, even if they were created before.
4 This employment share is consistent with that coming from other sources. For instance, the Structure, Consciousness and Class Biography Survey (Encuesta de Estructura, Conciencia y Biografia e Clase, ECBC), carried out in 1991, shows that the employment share of private firms with 1,000 employees or more was 10%. The temporary employment share of 20% in our sample is in contrast with the overall 33% given by the Labour Force Survey. However, it is well known that large firms use fixed-term contracts to a lesser extent. Again, the ECBC survey shows that in 1991 less than 20% of total employees in public and private large firms (those with 1,000 employees or more) had temporary contracts, and that proportion declined with increased firm size (for instance, it was above 40% for firms with less than ten employees).
5 In fact, under the Employment Promotion Programme of 1984, some fixed-term contracts can be renewed for up to three years. After that period of time on a fixed-term contract in the same firm, continued employment meant that the worker became permanent. In the case of termination of the contract, the same worker could not be re-employed by the firm under the same sort of employment relationship until one year had passed.
Anderson, P. and Meyer, B. (1994). 'The Extent and Consequences of Job Turnover', Brookings Papers on Economic Activity. Microeconomics, 177-248.
Bentolila, S. and Doldado J. (1994). 'Labour Flexibility and Wages: Lessons from Spain', Economic Policy, 18, 55-99.
Blanchard, O.J. and Diamond, P. (1990). 'The Cyclical Behavior of Gross Flows in the US Labor Market', Brookings Papers on Economic Activity, 2, 85-155.
Blanchflower, D. and Burgess, S. (1993). 'Job Creation and Job Destruction in Britain: 1980-90', Discussion Paper No. 912, Centre for Economic Policy Research, London.
Boeri, T. and Cramer, U. (1992). 'Employment Growth, Incumbents and Entrants', International Journal of Industrial Organization, 10, 545-65.
Booth, A. (1996). 'An Analysis of Firing Costs and their Implications for Unemployment Policy', in G. de la Dehesa and D.J. Shower (eds), Unemployment Policy: Government Options for the Labor Market, CEPR/CUP volume, London.
Contini, B., Gavosto, A., Revelli, R. and Sestito, P. (1992). 'Creazione e Distruzione di Posti di Lavoro in Italia', Temi di Discussione 177, Banca d'Italia.
Contini, B., Pacelli, L., Filippi, M., Lioni, G. and Revelli, R. (1995). A Study on Job Creation and Job Destruction in Europe, Study for the Commission of the European Communities.
Davis, S. and Haltiwanger, J. (1990). 'Gross Job Creation and Destruction: Microeconomic Evidence and Macroeconomic Implications', in S. Fischer and O. Blanchard (eds), NBER Macroeconomics Annual 1990, MIT Press, Cambridge, MA.
Davis, S. and Haltiwanger, J. (1992). 'Gross Job Creation, Gross Job Destruction and Employment Reallocation', Quarterly Journal of Economics, 107, 819-63.
Davis, S., Haltiwanger, J. and Schuh, S. (1994). 'Small Business and Job Creation: Dissecting the Myth and Reassesing the Facts', Business Economics, 29, 13-21.
Dolado, J.J. and Gomez, R. (1995). 'Creacion y destruccion de empleo en el sector privado manufacturero espanol: un analisis descriptivo', Investigaciones Economicas, 19, 371-93.
Dunne, T., Roberts, M. and Samuelson, L. (1989). 'The Growth and Failure of US Manufacturing Plants', Quarterly Journal of Economics, 54, 671-98.
Hamermesh, D. (1993). Labor Demand, Princeton University Press, Princeton, NJ.
Hamermesh, D. (1995). 'Labor Demand and the Source of Adjustment Costs', Economic Journal, 105, 620-34.
Hamermesh, D., Hassink, W. and van Ours, J. (1994). 'New Facts about Factor-Demand Dynamics: Employments, Jobs, and Workers', Working Paper No. 4625, National Bureau of Economic Research, Cambridge, MA.
Konings, J. (1995). 'Gross Job Creation and Gross Job Destruction in the UK Manufacturing Sector', Oxford Bulletin of Economics and Statistics, 57, 5-24.
Lagarde, S., Maurin, E. and Torelli, C. (1994). 'Creations et supressions d'emplois en France. Un etude de la periode 1984-1992', Economie et Prevision, 113/114, 67-88.
Lane, J., Stevens, D. and Burgess, S. (1996). 'Worker and Job Flows', Economics Letters, 51, 109-13.
Leonard, J. (1987). 'In the Wrong Place at the Wrong Time: the Extent of Frictional and Structural Unemployment', 141-63, in K. Lang and J. Leonard (eds), Unemployment and the Structure of Labor Markets, Basil Blackwell, Oxford.
OECD (1994). Employment Outlook, Paris.
OECD (1996). Employment Outlook, Paris.
Pissarides, C. (1986). 'Unemployment and Vacancies in Britain', Economic Policy, 3, 499-559.
|Printer friendly Cite/link Email Feedback|
|Author:||Serrano, Carlos Garcia|
|Publication:||Oxford Economic Papers|
|Date:||Oct 1, 1998|
|Previous Article:||Active labour market policies and job tenure.|
|Next Article:||Efficiency wages, trade unions, and employment.|