1) what are the strength and weakness of this article below

Does age matter for employability? A field experiment on ageism in the Swedish labour market

I. Introduction

Economists have conducted research on gender and racial discrimination in different markets for some time. Recently, discrimination against homosexual individuals has also gained increased attention. However, ageism, also called age discrimination, is relatively unexplored in economics. The purpose of this study is to contribute to this area of research by investigating whether people are discriminated against in the hiring process based on their age in Sweden. We conducted a field experiment in which we sent two matched applications from fictitious individuals, one from a younger male applicant and one from an older male applicant, to employers in two different industries (restaurant and sales). We then examined Whether the number of callbacks received from employers was related to the applicant’s age. The findings of this study provide strong evidence of ageism in the hiring process in the Swedish labour market.

These results support the findings of a handful field experiments that indicate the existence of ageism in the hiring process in other countries. Specifically, they are consistent with the findings of Bendick et al. (1996, 1999) in the United States; Riach and Rich (2006) in France; Riach and Rich (2007a) and Albert et al. (2011) in Spain; and Riach and Rich (2007b) in the United Kingdom. Never- theless, this study is the first to examine ageism in the Swedish labour market.

Sweden is an attractive outlet for conducting an investigation on ageism because a law prohibiting ageism was implemented in the labour market in2009. Sweden was the last country in the European Union to enact such a law. It is, therefore, interesting to investigate, especially from a policy perspective, whether this newly applied law prevents employers from discriminating against people on the basis of their age.

II. Methodology

We sent two applications, one from a fictitious 31year-old male applicant and one from a fictitious 46year-old male applicant, to employers with available jobs for restaurant workers and sales assistants. One reason for choosing these occupations was that the labour market demand in these areas was sufficiently high. Another reason was that we wanted occupations that did not require higher education. The ages were chosen because we wanted to study discrimination at a relatively young age. Vacant jobs were found on the website of the Swedish Public Employment Service, which is the main channel for job searches in Sweden.¹ We only applied to companies that accepted applications through email. Only a few employers wanted the applications to be submitted through regular mail; they were excluded from this study.

We applied to 466 available jobs in the selected occupations from the beginning of January 2011 through the middle of March 2011. Upon review of the job advertisements, we recorded the date of the job opening, name of the firm or company, city, email address and telephone number, as well as whether the job was temporary or full-time. The employers’ responses to the applications were recorded as a dichotomous outcome variable. Whenever employers invited the applicants for an interview or offered them a job, it was recorded as a callback. Invitations to interviews or job offers were promptly and politely declined to minimize the inconvenience to the employers.

Given that two applications were sent to each employer; they were closely but not completely matched to avoid being exposed. They were alternated between the younger and the older applicants to control for unknown and unobservable effects. Two of the most common Swedish distinctive male names, Anders and Karl, and two of the most common Swedish surnames, Johansson and Lindgren, were used in the applications.² These names are typical Swedish male names that cannot be mistaken for foreign names or female names. We used two real postal addresses in central Stockholm to which employers could send their written correspondences. Furthermore, two email accounts were created, and two telephone numbers with voicemail boxes were registered.³

The applications consisted of three parts. First, there was a message in the email that simply stated that the applicant was applying for the vacant job and that an application letter and a résumé were attached. The second part was an application letter that described the applicant in a narrative form. It began with the name of the applicant and a full correspondence address. The letter continued with the applicant stating that he had seen the announcement for the job opening and was interested in obtaining the position. The letter continued with information about the applicant’s age and work experience. The younger applicant was reportedly 31 years of age, whereas the older applicant was reportedly 46years of age. The letter stated that the applicant had 10 years of work experience in the specific occupation. The letter also indicated that the applicant had no difficulties collaborating with other people, liked working in the specific profession and spoke Swedish and English fluently. In one of the paragraphs, the applicant offered some information about his family, activities and hobbies. The applicant was reportedly married, but had no children. The third and final part of the application was a short résumé. It included the applicant’s contact information, date of birth, previous employments, education and language skills. All these information were carefully matched between the two applications and altered for the two different targeted occupations.

In addition to age, one other fact differed between the younger and the older applicants’ résumé in order´ to adjust for the 15 years of age difference. We wrote that the older applicant had been employed by the Swedish Armed Forces for 15 years and achieved the rank of lieutenant prior to his working as a sales assistant or restaurant worker. Otherwise, the younger and the older applicants had the exact same work experience in the occupations for which they were searching for a job.

III. Results

Table 1 summarizes the data and the results of this study. We computed the percentage of employers who did not respond (i.e. invite for an interview or offer a job) to either of the applicants (Neither), employers who responded positively to both of the applicants (Both), employers who only responded positively to the younger 31-year-old applicant (Younger) and employers who only responded positively to the older 46-year-old applicant (Older). The results are presented for the total sample, by occupation, for metropolitan areas and nonmetropolitan areas, for full-time and part-time jobs and for permanent and

temporary jobs.⁴

A total of 466 employers were sent two matched applications. As shown in the first row in Table 1, there were 42 (mean callback rate = 0.090, SD = 0.287) callbacks for the younger applicant and 13 (mean callback rate = 0.028, SD = 0.165) callbacks for the older applicant. That is, the younger applicant was 3 times more likely than the older applicant to receive an invitation for a job interview or a job offer from an employer. This difference was statistically significant, t(465) = 4.750, p , 0.001.

What is of interest in Table 1, however, is not to calculate the mean callback rates, but to examine whether there is symmetry between the number of employers who only contacted the younger applicant and the number of employers who only contacted the older applicant. If we let s_Y equal the probability that the employers only contacted the younger applicant and s_E equal the probability that the employers only contacted the older applicant, then the null hypothesis of symmetric treatment is

s_Y 1

¼ s_Eþs_Y 2

Regardless of how the data are organized, the p-values in Table 1, which were generated from a two-sided binomial sign test, show that we can reject the null hypothesis of symmetry at a = 0.001 in most cases. Therefore, the older applicant received significantly fewer responses from employers than the younger applicant, regardless of job characteristics.

IV. Conclusions

This article has provided strong evidence of ageism in the first stage of the hiring process in Sweden. The results suggest that ageism in the labour market is a serious problem. The extent of the discrimination was found to be higher than that reported in other similar studies examining different types of discrimination in the Swedish labour market. For example, Carlsson and Rooth (2007) found that a native Swedish applicant received 50% more callbacks from employers than an Arab/Muslim applicant. Furthermore, Ahmed et al. (2011) reported that a heterosexual males applicant received 14% more callbacks than a homosexual male applicant. They also found that a heterosexual female applicant received 22% more callbacks than a lesbian female. Similarly, Rooth (2009) stated that a normal weight person received 22% more callbacks than an obese person. These findings should be compared with the finding in this study that a 31-year-old applicant received 223% more callbacks than a 46-year-old applicant.

It is also worth emphasizing that a high degree of discrimination was found for people as young as 46, indicating that ageism is present at relatively early ages. The findings of this study suggest that there are

reasons to pay much more attention to the problems that people face because of their age in the labour market and other marketplaces than there has been in the past.