Directorates
·	Director General's Office
·	Data Development
·	Strategic Evaluations
·	Partnership and Continuous Evaluations
Publications
·	Evaluation Reports
·	Evaluation Briefs
·	Labour Market Development Agreement Series
·	Lessons Learned Series
·	Evaluation Tool Kit Series
·	Evaluation Forum Newsletter
·	From Theory To Practice

• Purpose
• Background
• Methodology and data
• Key findings
• Conclusions and issues for policy

Purpose

This briefing summarises an evaluation study that examines the relationship between the type of layoff and the way in which the UI program is used. This analysis has the unique feature of also considering the demand side of the labour market in interpreting the relationship between layoffs and usage of the UI system. In particular, the study has three objectives: 

1. To document the extent of temporary layoffs in the Canadian labour market;
2. To examine the relationship between temporary layoffs and repeat use of the UI program; and
3. To examine the number of weeks that laid-off individuals collect UI benefits according to their expectation of being recalled by a former employer.

Background

The current debate surrounding the Canadian Unemployment Insurance (UI) program has often focused on the interaction between UI and the supply side of the labour market almost to the complete exclusion of any issues associated with the demand side. The challenges facing the reform of UI are often phrased in terms of supply-side issues such as the "work disincentives" created or as outright "abuse" by claimants, but rarely in terms of demand-side issues such as the incentives faced by firms in their hiring and lay-off decisions. There are several reasons for this imbalance, the most important of which are probably related to the type of data available to researchers and policy analysts. Most of the data available to analyse the impact of UI on the labour market are based upon surveys of households. A great deal of information is collected on the behaviour of individuals, and very little on the firms that employ them.

The major labour demand issue concerns the consequences of less than perfectly "experience rated" UI premiums. A UI program is said to be experience-rated when the taxes that individual workers and firms pay vary according to the amount of benefits they are responsible for, being higher for those firms causing extensive use of UI and lower for others. The Canadian UI program is not experience-rated at all, and analysts have suggested that this implies the program offers an implicit subsidy to firms that will magnify the extent of layoffs, particularly temporary layoffs. Layoffs are one way of adjusting to a period of depressed sales, but others include changes in hours, and wages, or production for inventories, or even more fundamental changes in the organisation of work, in the job tasks and job skills of workers, and the degree of work-sharing. Basically, firms rely more on temporary layoffs and less on other mechanisms because they can shift part of the adjustment cost, through UI, to the public purse.

The definition of temporary layoffs used in the present study is based on the expectation that a laid off worker has (at the time of the layoff) of being recalled. Laid off workers have beliefs of being recalled that are held with varying degrees of certainty reflecting their understanding of the firm's intentions. It is these beliefs that influence their behaviour and, in particular, their use of the UI program.

Extensive repeat participation in the UI program has often been interpreted as a problem of labour supply. The availability and generosity of UI are thought to discourage individual claimants from undertaking adjustments that would increase their chances of obtaining stable long-term employment, or to encourage labour force participation for the sole purpose of qualifying for benefits. However, if temporary layoffs are an important element of the way in which the labour market functions, then interpretations of extensive repeat UI use based solely on supply side considerations are at best misleading. Individuals may repeatedly cycle between UI and employment with the same employer. The lack of experience rating may therefore have consequences for the way in which the UI program is used beyond its impact on the absolute number of claims generated.

Recall is always just a possibility and never an absolute certainty. Firms may, at the time of a lay-off, expect to recall an employee, but they may also have over-estimated growth in demand that would necessitate the recall, or in general their capacity to turn things around. Similarly they could make the contrary errors by laying workers off with no expectation of recall, but discovering that a downturn was only temporary and then attempt to recall them. Claimants expecting to be recalled have less of a tendency to search for a new job, and those whose expectations ultimately prove to be incorrect may as a result spend a very long period unemployed. The amount of benefits collected is in this way determined by both the decisions of workers and of their firms.

This study attempts to bring together both the labour supply and labour demand sides and focuses on how the use of the UI program is jointly influenced by the worker and employer decisions.

Methodology and data

The study uses information from the Record of Employment (ROE) and individual tax files to construct a measure of the number of temporary and permanent separations that is based solely upon actual outcomes. The total number of ROEs issued gives a measure of the number of separations occurring in the economy. A temporary separation is defined to be one in which the individual had some employment earnings from the same employer in the year following the separation. When an ROE is issued the employer is asked to indicate if it is expected that the employee will return to work, and the expected date of recall if any. This is the source of information on expected temporary layoffs used here. The first year of available recall information is 1986, and the data extends part way through 1992.

Since the focus of this study is the relationship between the type of separation and the UI program, ROE information is brought together with administrative data on UI based on a one-in-ten sample of claimants receiving UI benefits at some point between 1986 and 1988.

Key findings

Extent of temporary layoffs

Temporary layoffs are an important dimension of how the labour market functions. During the years 1986-88, almost 60 percent of all laid-off workers were recalled by their employers, but even a greater fraction, about 80 percent, expected to be recalled at the time of layoff.

Recall expectations are often not fulfilled. About 20 percent of those laid-off with a definite date of recall are ultimately not recalled. When a recall is expected, but no date is specified, closer to 40 percent of laid-off workers are not recalled.

On the other hand, 25 percent of laid-off workers who do not expect to be recalled end up returning to work for the employer that laid them off.

Temporary layoffs and repeat UI use

Repeat use of the UI program is considerable, but it should not be interpreted solely as resulting from the choices made by workers. Extensive dependence on UI is associated with a cycling from insured unemployment back to employment with a previous employer.

At any point in time, as many as 80 percent of UI claimants have collected UI in the past with as many as 40 percent experiencing their fifth or greater claim.

Over 40 percent of those with 5 or more claims over the 12 year period of 1978-89, supported their claims with employment from three or fewer different employers. The layoff and hiring decisions of firms are thus an important aspect of this cycle of dependence on UI.

Recall expectations and UI duration

The stronger the recall expectation, the lower the number of weeks of benefits collected during a claim. Over the period 1986-88, those expecting to be recalled who have a definite date (about 9 percent of laid-off claimants) collect only 16 weeks of benefits on average; those expecting recall without a definite date (68 percent of claimants) collect an average of 24 weeks of benefits; and those laid off with no expectation of recall (about 13 percent of laid-off claimants) collect 28 weeks of benefits on average.

Errors in recall expectations contribute to much longer periods of benefit receipt. Those expecting to be recalled but who ultimately are not represent about 30 percent of all laid-off claimants and account for about 35 percent of all benefits paid. Those laid off with strong expectations of recall and who, in fact, were ultimately recalled, collected an average of only 14 weeks of benefits, while those not recalled collected almost 23 weeks of benefits.

The supply-side determinants of benefit duration are distinctly secondary in their influence on the length of benefit receipt. The influence of recall expectations dwarfs the influence of any other factor including occupation, industry, province, age, or job tenure.

The chances of leaving UI in order to take up a job increase markedly as the exhaustion of benefits is approached, rising from about 10 percent to over 30 percent in the last few weeks of an entitlement. This is due not only to a rise in the chance of finding a new job, but also in the chance of being recalled. While some UI claimants appear to search more intensively for a new job as benefit exhaustion looms or become more willing to accept any job offer, some firms also appear to time their recall decisions with the benefit entitlement of their laid-off employees.

Conclusions and issues for policy

The major message of this study is that both the demand and supply side of the labour market matter in determining the way UI is used. A new measure of temporary layoffs has been introduced that is based on an individual's expectation of being recalled. Expected temporary layoffs are a very important element in the way the labour market functions, representing as much as 80 percent of all layoffs. This measure also alerts analysts to the existence of a previously neglected group of unemployed: those initially expecting recall whose expectations are not ultimately fulfilled. Temporary layoffs are also an important part of the explanation for extensive repeat use of the UI program, as well as of the number of weeks of benefits collected during any given claim. To the extent that the lack of experience-rating induces a greater volume of temporary layoffs than there would otherwise be, these findings can be thought of as some of the consequences.

In general, these findings offer a deeper understanding of the way the labour market functions and how it interacts with UI. They also raise at least three issues for the conduct of policy. The first deals with recommendations for mandatory notice before individual or mass layoffs. Such recommendations are often based upon analyses that make a sharp distinction between temporary and permanent separations. The notion of expected temporary layoffs introduced in this study suggests that reality is more complex, and that the target group for such a policy (those permanently laid off with no expectation of recall at the time of layoff) is a distinct minority. Those individuals laid off with an expectation of recall that ultimately proves incorrect are more numerous, but are not aided by such policies. There is a need to encourage a revelation of recall intentions from firms that are as accurate as possible. Advance notification does little in this regard and it may actually exacerbate the problem.

The results may also contribute to the reconsideration of the way the UI program is financed. Some analysts have argued that the program should be fully experience rated, that is that each firm should pay a premium rate that is related to the amount of benefits it is expected to be responsible for; others have argued that experience rating should be at the industry level.

This study, however, raises the possibility of a slightly different and milder form of experience rating. The finding that the probability of being recalled seems to increase sharply as the exhaustion of individual benefit entitlement approaches suggests that the incentives for employers to recall workers could be addressed by increasing the cost to the firm of long-term unemployment. Benefits paid to the long-term unemployed (that is, beyond six or seven months of unemployment) could be charged directly to the firm initiating the layoff, thereby encouraging firms to recall workers earlier. In the case of permanent layoffs or in the case of expected temporary layoffs that lead to permanent layoffs, firms would be encouraged to get more involved in the job counselling and skill needs of their employees. In this way the UI premium would change from being a "tax on jobs" to a "tax on unemployment."

The results, however, also establish challenges for the reform of the benefit structure of UI. The distinction is often drawn between a "passive" and an "active" UI program. The former offers income support during periods of unemployment, while the latter also requires some sort of obligation from claimants to undertake adjustments that are deemed to improve their future employability. An active program seeks to promote mobility of all types in the hope of encouraging stable patterns of employment and hence reducing reliance on UI. The Canadian UI program contains elements of both active and passive support. A very high fraction of laid-off individuals have a recall expectation. Those with an expectation of recall would be less inclined to participate in programs that are designed to break the bond between claimants and their previous employers. More generally, an active program must either correct or change recall expectations. If a job counselling program, for example, is to be effective it must address not only an individual's ability to find a new job, but also his or her belief in being recalled by a previous employer. Such a program may need to encourage individuals to revise their expectations of recall if necessary.

Further, there are two possible target groups for such a program: those with no expectation of recall, and those with an expectation that is unrealistic (or that will ultimately prove to be incorrect). Those with no expectation of recall, who are prone to receive UI for the longest length of time might be required to enrol in a job counselling program very early in a UI spell. Those with an expectation of recall but without a definite date might be required to enrol in such a program much later in a spell. Letting a longer time elapse permits a type of self-selection to occur. Those with a correct expectation of recall are most likely to have shorter UI spells. The entrance requirement for the program should be set long enough to let these individuals weed themselves out of possible participation, and thereby effectively target those individuals likely to hold incorrect recall expectations.

These are examples of the kinds of issues that could be debated. The fact that they have not been is a testament to the narrowness of a supply side focus on the labour market, a focus that the data presented in this study has hopefully served to broaden.