El origen real de los valores morales

The method just outlined relies on the central assumption that we observe - and thus can match on - all those differences between the various treatment groups that are likely to affect their outcomes.

The plausibility of the CIA should be discussed in relation to the richness of the available dataset as well as the selection process into the Swedish programmes. To this end it may be useful to separately consider:

(1) the decision between waiting further in open unemployment or joining a (i.e. any) programme;

(2) the decision to choose one specific programme among the available ones.

Figure 3.2 should help clarify the discussion that follows by highlighting the agents - the unemployed job-seeker, his caseworker and the local conditions prevail­ ing at his employment office - whose interactions determine the outcome of the se­ lection process (i.e. whether an individual joins a programme and if yes which one), as well as how these respective influences are captured in the available data.

As to decision (1), we need to control for all variables that, conditional on having spent a given amount of time in unemployment, influence both the decision to join a programme as well as potential future labour market performance were such decision to be postponed further. Since this was the basis for the analyses in the previous chap­ ter and was extensively discussed in Section II.5.3, here we only summarise the main issues. The decision between waiting further in open unemployment or joining a (i.e. any) programme appears to be driven by the individual’s subjective likelihood of em-

Figure 3.2 Selection process into the Swedish programmes and key available regres­ sors

1^^-time unemployed Elapsed unempl. duration

Pre-unemployment wage Part-time unemployment

Entitlement status Running out o f benefits

Y Employment history Subjective likelihood of employment Age Gender Part-time unemployment Specific & general education

Job-specific experience Occupation sought Citizenship...

/

\

Potential returns from participation Potential costs of participation Job seeker

Caseworker _{Local conditions}

selection into the programmes

Subjective overall evaluation of job­ seeker’s situation and character

Part-time employed Looking fo r part-time job

Willing to move Able to take jo b at once

In need o f guidance Difficult to place

Offered a programme I

County Month

At the individual’s office and that time:

Local programme rate ( ‘programme capacity’)

Local offer rate ( ‘waiting list’)

Local programme mix

ployment (Harkman, 2000, as reported in Carling and Richardson, 2001), which could in turn be proxied by several pieces of information characterising the recent

employment history of the individuals (in particular, elapsed unemployment duration in their first unemployment spell, entitlement status, and the additional piece of in­ formation in terms of pre-unemployment wage - a summary statistic of the worker’s past labour market situation). Similarly, we have controlled for factors relating both to employment prospects and either to potential returns from programme participation or affecting the opportunity cost or psychological cost of participation (age, gender, previous stock of human capital in terms of both specific and general education and job-specific experience, occupation being sought, citizenship, part-time unemploy­

ment status). Additional useful information allowing us to capture caseworker selec­ tion relates to the officials’ own subjective, synthetic and evolving overall evaluation of the situation and character of their unemployed client, summarising individual traits that are potential indicators of unobserved heterogeneity.

A possible source of violation of the ‘selection on observables’ assumption would be the presence of hidden job offers, that is if individuals waiting longer have decided to do so because they know they will be hired shortly. This would however not con­ stitute a serious problem if the typical time span between job offer and de-registration from the unemployment office is not too long.

Turning now to decision (2), i.e. the selection mechanism into the various pro­ grammes, the CIA requires the evaluator to have access to all the variables that influ­ ence both the choice between the programmes as well as potential future outcomes that would occur had the individual chosen an alternative progranune. Note that all our individuals have access to the same choice set, the only relevant recommendation being the one requiring a certain length of the unemployment period prior to enrol­ ment; benefit renewability rules and individual compensation while on the pro­ grammes are similarly comparable across programmes.

Harkman (2000) finds that while individual self-selection into different pro­ grammes is likely to be a minor issue in Sweden (unemployed workers tend to value the various programmes equally), the caseworkers do seem to have clear ideas about which type of programme is suitable for their clients, based on individual characteris­ tics. Since the relevant decision-maker thus appears to be the caseworker, the only

issue we need to focus on is whether he acts upon information which is unobserved to us and correlated with labour market outcomes. We do however observe not only im­ portant characteristics of the unemployed client, but also the caseworker’s own sub­ jective and synthetic evaluation of the overall situation and needs of service of his

unemployed client as described above. In a sense, the caseworker reveals, updates and records in the data a synthetic appraisal of various factors, including some which may have been originally unobserved to us. Our assumption then translates into the requirement that conditional on all this information, programme assignment is unre­ lated to outcomes; caseworkers or employment offices act idiosyncratically given worker characteristics (based e.g. on their preferences, incentives, experiences, col­ leagues’ opinions). Carling and Richardson (2001), who carefully examine the factors that determine in which programme the job seeker ends up into, do in fact provide reassuring evidence that the administrative selection process appears to be unrelated to the outcome.

Finally, in addition to county indicators, a set of local indicators at the individual’s municipality / employment office level over time have been included to further con­ trol for the possibility that individual joining decisions and/or office-specific pro­ gramme selection criteria may be based on local unobserved characteristics in turn correlated with individuals’ potential labour market performance. In addition to the local ‘programme-rate’ (the share of registered unemployed job-seekers participating in any programme) and the local ‘offer-rate’ (the proportion of unemployed workers who have been offered a programme out of all openly unemployed) used in the previ­ ous chapter, we have constructed a series of single programme ratios, reflecting the programme mix at that office and at that time.

III.5 Empirical findings

In this section, differential programme effects and the effect of joining a given pro­ gramme vis-à-vis waiting longer in unemployment are assessed in relation to two im-

portant outcomes: individuals’ employm ent rates over time and the probability o f be­

ing in a compensated unemployment spell over time.

In document El enfoque histórico en la filosofía de Nietzsche, antecedentes de la Genealogía de la moral (página 94-110)