Abstract
BACKGROUND:
The UN Convention on the Rights of Persons with Disabilities (CRPD) assumes that persons with disabilities have similar rights, motivations to work and personal values as those without disabilities.
OBJECTIVE:
The article examines the corroboration between this assumption and real-life facts to better understand the importance of labor-oriented values in people with disabilities.
METHODS:
We tested the relationship between human values, employment and wages among Israelis with disabilities who cope with prejudice, negative attitudes and a lack of accessible workplaces in comparison to Israelis without disabilities.
RESULTS:
We found that the effect of labor-oriented values on employment status is 70% higher among people with disabilities than among those without disabilities. Furthermore, persons with disabilities ranked power and achievement as important values related to employment, but these values were not included in the considerations of persons without disabilities.
CONCLUSIONS:
These results highlight the importance of labor-oriented values for people with disabilities to overcome challenges in the labor market. Our findings suggest that rehabilitation policies would benefit from identifying personal human values of people with disabilities at an early stage of their career.
Introduction
Societal views on the rights of persons with disabilities have changed over the last few decades. The first change relates to the transition from the medicalization of persons with disabilities to a social functional model, signaling that changes in their status require social inclusion [1]. However, the most visible change involves the development of the United Nations Convention on the Rights of Persons with Disabilities (CRPD) and its purpose of guiding policies and practices [2]. The adoption of the CRPD in 2006 marks a paradigm change toward human rights for persons with disabilities [3]. Particularly, it significantly transitions from “objective needs to subjective rights and the promotion of quality of life” [4]. According to Article 27 [5] of the CRPD, “States Parties recognize the right of persons with disabilities to work, on an equal basis with others; this includes the right to the opportunity to gain a living by work freely chosen or accepted in a labor market and work environment that is open, inclusive and accessible to persons with disabilities.”
Therefore, Article 27 marks a change from people with disabilities as passive recipients of care to people who are expected to exercise their self-determination and choices. The human rights model in Article 27 defined the right to work of persons with disabilities as equal with others’ right to work [5]. This includes the opportunity for persons with disabilities to earn a living by freely choosing employment in an open, accessible and inclusive labor market [6]. The responsibilities are evident: states must create non-discriminatory, accessible and reasonable accommodations for work, and people with disabilities have the right to conceive their own preferences and make their own choices regarding employment. The human rights model, and particularly the CRPD, emphasizes a person-centered approach and recognizes that persons with disabilities have self-determination and are entitled to make choices and decisions about their lives [7].
Across OECD countries, only 40% of people with disabilities are employed, compared to 75% of people without a disability (OECD, 2010). The employment rates among Israelis with and without disabilities are quite similar—45% and 71%, respectively [8]. A survey in the UK demonstrated that persons with disabilities want to work, but one-third of employed people with disabilities and two-thirds of unemployed people with disabilities reported being limited in the amount or type of work they could do [9]. Although significant differences exist in employment rates and in the income of people with and without disabilities, there is almost no evidence that persons with disabilities have similar or different personal values regarding employment and earnings than those without a disability [6]. In fact, the CRPD’s hidden assumption is that persons with disabilities have human and personal values similar to those of people without disabilities and that the states have to set the conditions to express their desires. Examining the values associated with employment is consonant with the language of the new convention, and it helps policy-makers understand the motivations of people with disabilities to work in the current labormarket.
One accepted theory for examining motivational characteristics is Schwartz’s theory of human values [10, 11]. Human values are cognitive representations of desires and goals that vary in importance as guiding principles in people’s lives [10]. Schwartz’s theory of human values is a cross-cultural, reliable and validated theory, and the literature found that this validity does not depend on how we measure values (explicitly, indirectly or by others); moreover, the literature shows that this theory applies to various populations around the world [11]. Our measurement was based on personal accounts evaluated via a questionnaire that identifies 10 human values across two major dimensions: self-enhancement versus self-transcendence and self-enhancement versus conservatism. Roccas et al. [12] show that human values demonstrate what people consider important—distinctively from the big five personality traits (Openness to Experience, Conscientiousness, Extraversion, Agreeableness, Neuroticism) that refer to what people are like. Previous studies have shown that the dimensions of self-transcendence and openness to change correlate with intrinsic, extrinsic, social and power-related work values [13, 14]. Therefore, the probability of working and the determination of wages plausibly reflect a higher value of labor-oriented human values such as self-direction, stimulation, achievement and more.
To reveal the correlation between labor-oriented values and the labor outcomes of people with disabilities in Israel, we compare individuals with and without disabilities. In order to work, people with disabilities must overcome structural limitations in the labor market, such as accessibility, stigma and discrimination. Moreover, the disability allowance in Israel is a means-tested benefit paid to people whose earning capacity is reduced due to their disability. Thus, employed individuals with disabilities are eligible for lower social benefits than unemployed individuals with disabilities. Past studies have shown that entitlement to a monthly allowance decreases the profitability of integrating into the workforce [15, 16]. Although Israeli law have dealt with the disincentive aspect of benefits by establishing a relatively generous phasing-out as a function of labor wages, our analysis will include the impact of benefits on household income.
Article 27 of the CRPD emphasizes the importance of listening to the perspectives of persons with disabilities regarding work. A previous study has shown that employed people with disabilities felt that work gave their lives meaning [17], especially if they consider their job meaningful [18]. Moreover, employed Israelis with disabilities were more integrated in leisure and civic activities than their non-employed counterparts [19]. This study explores the link between persons with disabilities’ human values and objective labor market outcomes, as compared to those without disabilities. In addition, the research explores the relationship between human values and both the probability to work and wages.
The empirical framework
Social benefits for people with disabilities in Israel
The State of Israel has comprehensive legislation for people with disabilities, including social allowances and monetary benefits. Among others, people with disabilities are eligible for a benefit from the Israeli National Insurance Institute for illnesses or injuries outside the workplace that damage their capacity to earn an income. On average, 200,000 people in Israel received a disability allowance each month.
Disability benefit shows a gradual decrease in correlation with increased income from work. Beyond approximately 2,000 NIS, the benefit decreases as work income increases (on average, the benefit decreases by 0.3 NIS for every additional 1 NIS income from work). This ensures that overall income (benefits and income from work) of benefit recipients who participate in the workforce will grow in correlation with increased income from work. Therefore, recipients of benefits have an incentive to work.
Another benefit is the tax credit for people with disabilities. In order to receive the tax credit, disabled individuals undergo an assessment (which is identical to the one that is performed by Social Security Institute) to discern medical disability to receive disability benefits. Tax benefits due to medical disability apply to income from work and from other sources (interest, rent, etc.). Tax benefits for income from work are eight times higher than tax benefits for income not from work.
Sources of information
Our study identified people with disabilities via the Israeli social security system’s database. Data on demographics, income and medical information were examined by the medical committee. Employment data in the National Insurance Institute (aka NII) were taken directly from the tax authorities at the end of the tax year, and includes information on anyone receiving wages subject to tax payments.
At the time of the research, the NII had updated employment data through 2013. This database also outlines the relationships between immediate relatives (parents, brothers and children) and between spouses (married or de facto). Furthermore, this database identifies people’s individual economic status in the years prior to examination.
The data collected by The European Research Infrastructure Organization, which maintains a database for research and education, allowed us to identify a sample of individuals without a disability within Israel’s general population. In the sixth cycle of the European Social Survey in 2012, a sample of 2,507 individuals of working age (15 years old and above) in Israel participated. Data in the survey includes demographic and economic information on the participants as well as information on attitudes, values and behaviors reflected in Schwartz’s model. This information also included the total income of all members of any given household and details only at the point ofexamination.
Population group
Using the NII’s database, we analyzed a sample group of benefits recipients, which likely includes people with disabilities who work and those who do not. To determine whether an individual can choose between income from work or from a benefit, we identified those who were of prime working age when their disability developed (men: 18–67; women: 18–62), were entitled to disability benefits between 2003 and 2012, had a permanent 40% (minimum) medical disability, and had worked for six months prior to the disability as employees or in a self-employed capacity. At the end of this stage, our sample included 56,327 individuals entitled to a disability allowance and 8,578 individuals eligible for benefits on income taxes due to disability.
In the second stage, we minimized the demographic and medical variance between the two groups, and we randomly chose a sample of 750 people from each group for participation in our survey on human values. To measure the human values of people with disabilities, we conducted a phone survey between August 2014 and January 2015 via the Surveys Unit at the NII. The survey was conducted in Hebrew or in Arabic (depends on the participants’ preference) and used the reliable and validated Schwartz questionnaire in both languages.A willingness to answer the telephone survey was considered consent to participate, and identifying variables were deleted after the survey was completed. The research was approved by the NII committee for Ethics in Research and Submission of Information, and the information gathered in no way affected the respondents’ benefits.
Approximately 51% of the individuals with disabilities who were approached during the survey agreed to participate. In Table 1, we highlight selected characteristics among survey participants compared to those who did not participate. As demonstrated, the differences in gender and employment status between people who participated in the survey and those who did not participate were insignificant. Similar results were found regarding age, ethnicity and medical disability.
People with disabilities: Characteristics of survey participants versus those who did not participate
People with disabilities: Characteristics of survey participants versus those who did not participate
Data from the European Research Infrastructure Organization database only reports the total income of all members of any given household—it does not distinguish individual income from overall household income—so we decided to focus only on individuals without a disability who were of prime working age (men: 18–67; women: 18–62) whose household income is based only on work. We identified these individuals based on two criteria: (1) the participant in the survey works, has no alternative income, and information on the amount of income exists; and (2) there are no other people in the household with any type of income. Therefore, only people without a disability either living alone or with an unemployed partner or children under the age of 15 were identified in our study. Additionally, the number of people in the household was limited to four to lower the likelihood that the household income includes unobserved elements. Altogether, 117 people fulfilled these conditions.
We created an additional group of participants in the survey who were of working age but not employed. Altogether, 648 people fulfilled this condition and inherently earned zero income from work. Selected characteristics of the surveyed individuals without a disability and the ratio to participants in the ESS database are presented in Table 2.
People without a disability: Characteristics of working and non-working participants
Our final stage included a pairing process based on several criteria 1 to overcome differences inoccupation linked to gender, ethnicity, medicine, age, childcare and work intensity. For individuals with disabilities who participated in the survey,the conditions in this pairing process are: (1) gender (men/women); (2) age (a difference of up to 10 years between people); (3) ethnicity (general Jews/ultra-Orthodox Jews/Israeli Arabs); (4) parents of children under the age of 18 (yes/no); (5) dominant medical disability (mental health/nervous system/internal organs disease or injury/loco-motional dysfunction/audiovisual dysfunction/other); (6) severity of disability (up to 10% variation); and (7) area or residence (according to the Bureau of Statistics). To create groups of equal size, a most similar “pair” was identified for each individual in the groups. After this stage, 484 participants (divided into 242 pairs) remained. Of the participants, 50% were people with disabilities who worked in December 2013, and 50% were people with disabilities who did not work in December2013.
The pairing of people without a disability was based on the following characteristics: (1) gender (men/women); (2) age (a gap of up to 10 years between people); (3) ethnicity (general Jews/ ultra-orthodox Jews/ Israeli Arabs); (4) parents to children under the age of 18 (yes/no). After this process, 107 pairs of people without a disability remained, half of whom were working at the time of the ESS survey, and half of whom were not working at the time of the survey.
Table 3 compares selected characteristics between the two groups: people with disabilities (from the survey we conducted) and people without a disability (from the ESS survey). This comparison only applied to working individuals for whom a “twin” was identified among the people who did not work. Our results show observable differences between the two groups, particularly regarding age and marital status. Thus, we created a subsample from within the two groups comprised of unmarried people aged 40 to 60. This aimed at reducing the difference between the groups. We tested these results (as discussed later) to examine the significance of these differences.
People with and without disabilities: Selected characteristics of working people
In accordance with Schwartz’s theory, human values were measured using a questionnaire. Participants were asked to rate all human values from 1 (does not resemble me at all) to 5 (resembles me greatly), but we were especially interested in the labor-oriented values of self-direction, stimulation, hedonism, power and achievement.
Aside from the human values, other personal characteristics were tested: ethnicity (general Jews/ultra-Orthodox Jews/Israeli Arabs), marital status (married/not married), academic degree (yes/other), monthly allowance (yes/no), first-degree relatives who work (sequential variables from 1 to 0), position at work prior to disability (self-employed/employee) and occupational seniority prior to disability (sequential variable). Those variables are popular in labor analysis and were tested to overcome statisticalartifacts.
To estimate the influence of personal characteristics on work prospects and wage, we adopted a number of methodologies to identify differences between working individuals and those who do not work. Likewise, we examined separately the relationship between each tested variable, the possibility to work and the size of the wage. This enabled us to examine the variance of personal characteristics between people with highly similar demographic information but a different work status.
We used two techniques of restricted models 2 . First, we used a logistic-regression test focused on factors affecting employment (without differentiating between wage-related factors and factors linked to the decision to work). We considered interfering factors for outliers.
To separately evaluate the marginal effect of each variable that explains employment and wage, we also adopted the Type 2 Tobit method. This method provides a simultaneous solution for the problem of choice (employment) and for evaluating the sequential variable (wage) while also considering the unobserved wage prediction as a result of deciding not to work [21].
Results
The relationship between human values and employment
The first test examined the average differences between similar individuals based on the differential method (Table 4). Within each group (people with disabilities and people without a disability), pairs were made in which one individual worked and one did not. A positive and significant correlation between human values and work are indicated, and some values displayed a difference between people with disabilities and people without a disability. We found that people with high self-direction, achievement, stimulation and hedonism are significantly more likely to work, with no difference between the individuals with disabilities and individuals without a disability. On the other hand, power’s value among people with disabilities has a positive and significant effect on the probability to work. This observation did not apply to people without a disability. Among people without a disability, the values of benevolence and universalism have a significant and positive effect on the probability to work. This observation did not apply to people with disabilities.
Difference-in-differences estimation: Pairs of working and non-working people with and without disabilities
Difference-in-differences estimation: Pairs of working and non-working people with and without disabilities
Echoing the findings of past research, a tertiary academic degree and being self-employed before the disability is positively and significantly correlated with employment of people with disabilities. We also found a significant negative correlation between employment, receiving disability allowance and having alternative sources of income (relevant only to people with disabilities). In addition, our results showed a negative correlation between marriage and employment among people without a disability.
To control for interfering variables and to identify the effect of human values, we applied regression tests. Using logistic regression, we examined the correlation between various factors and participation in the workforce among people with and without disabilities. The variable had a value of 1 for people who worked and 0 for people who did not work (Table 5). The results demonstrate that self-direction has a positive significant link with employment for people with and without disabilities. Although the significance level of power is only 6%, the employment of people with disabilities seemed to correlate with more than just one value. Interestingly, this value’s correlation with employment is higher for people with disabilities.
Logistic regression test: The relation between personal characteristics and the probability to work among people with and without disabilities
Beyond these results, a positive correlation between independent work, the number of working relatives and the probability to work can be found among people with disabilities. Likewise, we found a strong negative correlation between the probability to work and being married. Our findings also showed that occupational seniority prior to disability positively correlated with the probability of working. However, this marginal probability decreases if an individual begins receiving a monthly allowance.
Based on the classic economic model, the probability to work depends on two factors: the wage offered in the workforce and the subjective reservation wage. To identify the relationship between human values, employment and wage, we applied a Type 2 Tobit regression test to separately examine the probability of working and wage while accounting for interfering variables. This test facilitated a simultaneous estimation that differentiated between elements that affect the decision to work and the wage, respectively.
Results of the regression test for the correlation of human values with employment and with wage among people with disabilities appear in Table 6. The probability of working among people with disabilities positively correlates with self-direction. However, wage does not correlate with this value.
Tobit regression test: The relation between personal characteristics and the probability to work and log-wage among people with disabilities
Tobit regression test: The relation between personal characteristics and the probability to work and log-wage among people with disabilities
The results also demonstrate that academic qualification and age significantly affect wage but not the probability of working. The findings also illustrate that age has an opposite effect among recipients of disability benefits. On the other hand, occupational seniority and self-employment prior to the disability’s onset appear linked to the probability of working but not to wage.
The results of the Type 2 Tobit regression test among people without a disability appear in Table 7 and show that self-direction positively correlates with the probability of working but does not influence wage. Comparing Tables 6 and 7 reveals that the correlation between self-direction and employment among people with disabilities is higher than that correlation among people without a disability: 0.69 + 0.25 for people with disabilities compared to 0.21 among people without a disability. Furthermore, the value achievement was not significant, and self-direction does not appear to be the only influence on the employment of people with disabilities.
Tobit regression test: The relation of personal characteristics and the probability to work and log-wage among people without a disability
Moreover, findings demonstrate that being married negatively correlates with the probability of working, but not with wage. On the other hand, results show that academic qualification is individually linked to wage but not to employment.
The findings above strengthen our assumption that human values more prominently affect people with disabilities compared to people without a disability. In light of the variance between age and marital status, we assessed whether these demographic factors affected the results. To answer this question, we applied another test for unmarried people aged 40 to 60 (Table 8). The results of this test supported our findings that self-direction significantly correlates with the probability of working and that its effect is 70% higher among people with disabilities than among people without a disability (0.7 as opposed to 0.41, respectively). The low number of observations explains the insignificance of the coefficient regarding academic qualification among people with disabilities.
Unmarried people aged 40 to 60: Tobit regression test for the relation between personal characteristics and the probability to work and log-wage among people with and without disabilities
Consonant with Article 27 of the CRPD and with the Disability Rights model, our research examines the association between perceived human values, employment and wage outcomes in Israel for persons with disabilities compared to persons without disabilities. Human values are measured according to Schwartz’s theory, and the research data are a combination of survey and administrative data.
By adopting several methodologies, we provide evidence that the effect of labor-oriented values on employment is 70% higher among people with disabilities than among people without a disability. Furthermore, persons with disabilities ranked power and achievement as important values in working, but these values were absent from considerations among persons without a disability. Additionally, our results show no correlation between values and wages, suggesting that human values only affect the internal motivation to work—not the salary, which is usually determined by the labor market.
Generalizing this finding has a practical implication for policy: raising the household income for people with disabilities is associated with integrating them into work. According to our findings, the probability of successfully integrating into the labor market for people with disabilities is higher for self-directed persons.
Employed persons with disabilities perceive employment as more desirable, probably because they have to cope with prejudices, negative attitudes, a lack of accessible workplaces and a discouraging corporate culture [23]. The latter seems to perceive employees with disabilities as challenging, requiring additional resources and efforts to accommodate them [24].
Given the structural limitations that people with disabilities are constrained to within the labor market [23], such as reserved attitudes toward employability, a lack of accessibility and integration in the workplace, our findings highlight the role of labor-oriented values—such as self-direction—for overcoming the labor market’s challenges. Notably, the underlying assumption of Article 27 of the CRPD is that persons with disabilities have an equal right to work and that the labor market and work environment must be inclusive and accessible. Our findings imply a testable examination for this principle: accessibility shall be at least enforced among a labor-oriented values population of individuals with disabilities.
Our research has important policy implications. Earlier studies aimed at matching traits with the meaning of work [15, 25]. By contrast, our findings show that human values are crucial for making policy decisions about how people with disabilities make choices regarding their employment. In particular, human values must be considered in any effort to examine the person–job fit and in any career-development counseling [26]. It appears that the real obstacles are structural and not hidden in their lack of interest or disinterest in employment.
Concerning practical policy recommendations, our results stress that employing more people with disabilities requires actions that strengthen the intrinsic importance of work to improve their subjective well-being. In particular, rehabilitation policies should account for the values of people with disabilities. It is important to understand the human values of people with disabilities early in their careers to perform the right diagnostic about the need to enhance characteristics that help their work career. In particular, we find that enhancing self-direction increases their probability of success in labor market integration. Performing this diagnostic early in the career increases their chance of integration.
Conflict of interest
None to report.
Footnotes
Acknowledgments
We are very grateful to Hester Fass, Michael Beenstock, Micha Mandel, Ram Fishman, Lilach Sagiv, Ariel Knafo, Roni Holler, Rachel Caplin and two anonymous reviewers for their help and valuable comments on this article. The research was supported and funded by the National Insurance Institute of Israel. The authors are responsible for any remaining errors or omissions. Due to privacy concerns, study materials are available only at NII’s research room.
Along the lines of the Statistic Method, as explained byCaliendo and Kopeining (20). We performed many regressions using different variables and chose the variables that performed the best.
Fisher (22, p. 42) stated that the accepted level of significance is not given ex-ante, and it depends on the particular research characteristics. In our case, the number of observations is relatively low, and each one has a higher weight because of the low number of observations. Consequently we considered 5% as a limit level for statistical significance, but we also report variables up to a 10% significance level.