table of contents new ceo for srdc · newsletter more attractive and easier for you to read, it has...

Table of Contents

New CEO for SRDC � 1

Making Work Pay SymposiumPresents Key Lessons forDeveloping Income Support Programs That Encourage Work � 1

SRDC Has a New Look � 3

SSP’s Impacts on the Distribution of Earnings � 6

Case Coordination and the Hard-to-Employ: An Initiative to Help the Long-Term Unemployed Move Into Work � 9

Institutional Factors Can Affect Policy Reforms � 13

SSP, Education, and Welfare Exits � 15

SRDC Wins New Contract for a Francophone Early Childhood Project � 16

Bulletin Board � 17

Making Work Pay SymposiumPresents Key Lessons forDeveloping Income SupportPrograms That Encourage WorkThe Making Work Pay Symposium, funded by Human Resources and SkillsDevelopment Canada (HRSDC) and organized by SRDC, was held in Ottawa onNovember 15 and 16, 2005. The aim of the symposium was to help policy-makers andprogram practitioners bring together research and practical experience on the use ofprograms designed to encourage work while providing income support. With that aim,the symposium heard about a number of programs in Canadian provinces that try to“make work pay.” In addition, they heard from a number of well-known academic

S O C I A L R E S E A R C H A N D D E M O N S T R A T I O N C O R P O R A T I O N

1

New CEO for SRDCIn July, John Greenwood will be stepping down from his position as Executive Directorafter almost 11 years with SRDC. The new Executive Director and Chief ExecutiveOfficer will be Jean-Pierre Voyer. For the past four years, Mr. Voyer has been theExecutive Director of the Policy Research Initiative, an organization attached to the PrivyCouncil Office and responsible for conducting research on cross-cutting social, economic,and environmental issues in support of the Government of Canada’s medium-term policyagenda. During his career, Mr. Voyer has held positions in the Department of Finance, theNational Union of Government Employees, the Regional and Economic DevelopmentSecretariat of the Privy Council Office, and the Economic Council of Canada. From 1994to 2000 he was Director General of the Applied Research Branch at Human ResourcesDevelopment Canada (HRDC), which covered a wide spectrum of labour market andsocial policy issues and provided direction and financial support for large surveys relevantto HRDC’s mandate.

Mr. Voyer has represented Canada on numerous occasions at the Organisation forEconomic Co-operation and Development (OECD) and at other international meetings. Hewas Chairman of the OECD Education, Labour and Social Affairs Committee from 1998to 2000. He currently sits on the research advisory committees of several research projectsand organizations, including the Social Sciences and Humanities Research Council ofCanada, the University of Ottawa, and Human Resources and Social Development Canada.He is also no stranger to SRDC: between 2000 and 2002 Mr. Voyer served as SRDC’sDeputy Executive Director. He holds a Master’s degree in Economics from Queen’sUniversity and an undergraduate degree in Economics from Université de Montréal.

E V I D E N C E F R O M S R D C ’ S S O C I A L E X P E R I M E N T S A N D R E S E A R C H

2

researchers who are using data from Canada’s mostfamous social policy experiment — the Self-SufficiencyProject (SSP). The research findings produced by SSPare available in many reports and working paperspublished by SRDC; consequently, this article focuses onthe presentations made by provincial representativesattending the symposium.

BackgroundThe dilemma of policies designed to fight poverty is thatgiving money to the poor can decrease their incentive tofind and maintain employment and to attain financialself-sufficiency through their work efforts. This isparticularly true if the jobs that they can obtain offer lowpay — in some circumstances less than the amount thatcan be received from income assistance. This concernhas led to a shift in policy interest toward the provisionof “in-work” benefits designed to make work financiallyattractive. SSP was a test of one such initiative. Otherexamples include the National Child Tax Benefit and,potentially, the Working Income Tax Benefit (WITB) thatwas proposed in the federal Finance Minister’s economicand fiscal update on the eve of the symposium. At theprovincial level, there are ongoing initiatives in manyprovinces, including British Columbia, Alberta,Saskatchewan, Newfoundland, Ontario, and Quebec.

Lessons learned The symposium discussed a number of lessons learnedabout making-work-pay programs. Such programs are

successful, widely accepted, and diverse, includingtraditional financial incentives such as earningssupplements and tax credits but also housing programsand supplementary health and drug card programs.However, provincial governments sometimes disagree onthe elements of making-work-pay strategies. Forexample, the British Columbia government haseliminated welfare earnings disregards, a policy thatallows welfare recipients to retain part of any earningsthey receive while on welfare without incurring a dollar-for-dollar reduction in benefits. BC believes this policypromotes welfare dependency by making it easier formany job-ready clients to remain on welfare. In contrast,Ontario believes that it is important that clients increasetheir earnings while receiving welfare as a first step intothe labour market. Consequently, Ontario has made itsearnings disregard policy more generous.

The symposium heard that benefit formulas should besimple so that intended beneficiaries can understand theadvantages offered by the program and what they have todo in order to receive those benefits. Some potentialbeneficiaries may fail to apply for benefits because theydo not fully understand the benefit formula and,consequently, perceive program participation as risky.The level of perceived risk will increase if the benefitformulas involve retrospective reconciliations that canresult in clients having to repay previously receivedbenefits if, for example, they are determined to haveaccumulated earnings above some threshold level in agiven year.

Poor communications strategy can hamper programparticipation by failing to explain the goals and potentialbenefits of the program. In addition, poorcommunications can transform a “good news” story, suchas increased benefits for the working poor into a “badnews” controversy, the symposium heard.

Inadequate resourcing of a program can also impede theattainment of the stated policy goals. For example,employment plans and job-search requirements may existonly on paper unless there are adequate resources toimplement them effectively. This disparity betweenpolicy aims and resources can create the illusion of anemployment-oriented policy in a department that is, inpractice, continuing mainly to provide income supportpayments with little in the way of effective ancillarysupports and services. In addition, paternalistic practices,such as relieving clients of responsibility for theirpersonal and financial decisions, can increasedependency, the symposium heard.

is published by the Social Research and DemonstrationCorporation (SRDC).

SRDC’s two-part mission is to help policy-makers andpractitioners identify and implement social policies and programsthat improve the well-being of all Canadians, with a specialconcern for the effects on the disadvantaged, and to raise thestandards of evidence that are used in assessing social policiesand programs.

Complimentary subscriptions to Learning What Works and copiesof SRDC publications are available free of charge by contactingSRDC at 55 Murray Street, Suite 400Ottawa, Ontario K1N 5M3, Canada.Tel.: (613) 237-4311 Fax: (613) 237-5045E-mail: [email protected]

Electronic versions are available at www.srdc.org.

ISSN 1496-8622


3

Spring 2006

Some provinces have chosen to subcontract servicedelivery to non-governmental agencies in either the for-profit or non-profit sector. These service deliveryorganizations can be successful in some cases butproblematic in others. Establishing processes fordetermining success, measuring performance, andproviding compensation linked to results can bechallenging.

Among the emerging priorities is an increased emphasison job-retention strategies for clients who use welfarefrequently and the development of supports for clientswith multiple barriers.

Provincial presentationsMost of these “lessons learned” came from a series ofpresentations by provincial governments on the secondday of the symposium. These presentations outlined thepolicies, accomplishments, and challenges of makingwork pay in Saskatchewan, British Columbia, Quebec,Newfoundland, Alberta, and Ontario.

Doug Scott, a Senior Policy Analyst at theSaskatchewan Department of Community Resourcesand Employment, characterized the Saskatchewanwelfare system in 1997 as administered through welfareoffices with face-to-face contact, heavy caseloads,limited case management, cursory employment plans,and an emphasis on benefit administration. By 2005 the

system had changed radically with a much greateremphasis on the provision of employment assistance,including intensive case planning, and theadministration of benefits through regional contactcentres and increasingly through the income tax system.In addition, Saskatchewan made a wider range ofbenefits available to the working poor. For example,provincial and national child tax benefits replacedwelfare child benefits, and the SaskatchewanEmployment Supplement gave earnings supplements tothe working poor. However, Scott also said that take-upof the supplement was slow to build due tocommunications strategies that too often left clientsconfused about the nature and purpose of the program.In 2005 Saskatchewan introduced rental subsidies thatare available to both welfare clientele and working poor.

Over the last 10 years the “employable” caseload inSaskatchewan has been reduced by half, but there hasbeen little change in the caseload of the not-fully-employable, “which points to the challenge that liesbefore us,” Scott concluded.

Robert Bruce, Manager of Research and EconomicAnalysis in the British Columbia Ministry ofEmployment and Income Assistance, said his provincehas implemented two major waves of welfare reforms,first in 1996, then in 2002. In January 1996 BritishColumbia introduced BC Benefits, a program thatplaced strong emphasis on employment and training for

SRDC Has a New LookAs you have noticed, SRDC’s newsletter Learning What Works has a new look! In an effort to make ournewsletter more attractive and easier for you to read, it has been redesigned. SRDC is in the process ofmaking improvements to several of its communications materials, which has recently included the redesign ofour corporate logo and our Web site.

After many years of having separate English and French logos, we now have a single bilingual logo thatincorporates the highly recognizable text style of the original SRDC and SRSA initialisms with a stylized redmaple leaf. As SRDC is a Canadian organization whose mission is in part “to help policy-makers andpractitioners identify social policies and programs that improve the well-being of all Canadians,” the redmaple leaf was a natural addition to the logo.

We have also recently launched a redesigned SRDC Web site. Next time you visit www.srdc.org, you willnotice a brand new look. In addition, our Web site has undergone several enhancements to help you navigatethe site more easily and find the information you need. For example the Web site now offers a powerful newPublications Search tool to help you find the SRDC publications most relevant to your work.


4

employable welfare recipients; initiated the FamilyBonus, which subsequently became the model for theNational Child Benefit; and implemented mandatorytraining for all youth aged 19 to 24 applying for incomeassistance as well as an optional training program forolder clients.

The 2002 reforms were, according to Bruce,“revolutionary in Canada and a huge cultural change forthe Ministry.” These reforms included mandatoryemployment plans for all employable clients,enforcement of sanctions, a mandatory three-week jobsearch for applicants prior to being admitted to thewelfare rolls, and a requirement that single parents seekwork once their youngest children reach 3 years of age.One controversial reform was the removal of earningexemptions for employable clients. Earnings exemptionsare usually thought to encourage employment, but theBC government felt they encouraged a long-termcombination of welfare and work. Finally, BC introduceda two-year time limit on the receipt of income assistancefor many employable clients. There are a large numberof exemptions to the time limits; however, voluntaryexits from welfare have been seen to increase as clientsapproach the two-year limit, which likely results from acombination of the effect of the impending loss ofbenefits and the priority the Ministry gives to providingservices to these clients, Bruce said.

Over the nine years during which these reforms tookplace, the employable caseload in BC has declined by113,000 cases (or 83 per cent) even though the definitionof “employable” has been expanded to include clientswith drug and alcohol problems, mild mental illnessproblems, or mild medical conditions. Substantialcaseload reductions have been seen across all familytypes. For example, Bruce described how the number ofcases represented by single parents with children under3 years of age had decreased by 43 per cent, even thoughthese clients were exempt from some of the policychanges such as time limits and mandatory job search. Inaddition, the composition of the current BC employablecaseload is approximately the same in terms of educationand work experience as it was prior to the reforms,indicating that clients with a range of characteristicsexperienced similar caseload reductions. Some “barriers”may not be barriers at all, Bruce said. However, he alsosaw increasing the employment opportunities available topersons with disabilities as a key future priority.

Serge Hamel, Director of Evaluation at the QuebecMinistry of Employment and Social Solidarity, revieweda number of Quebec programs that have been introducedto make work pay over the past 20 years. In 1988 theParental Wage Support Program (APPORT) wasintroduced to provide monthly benefits to families withchildren as a supplement to earned income. Potentialbeneficiaries found the complex program difficult tounderstand, which limited the take-up rate and raisedconcerns over fairness. For example, monthly benefitswere based on projected income for the year but werelater reconciled with actual income. Consequently somefamilies had to reimburse benefits. An evaluation of theprogram found that it increased earnings and the numberof weeks worked by participants and decreased benefitsreceived from Employment Insurance, but it increasedthe total amount of income support received.

Following consultations with SRDC, Quebec introduceda pilot project in 2001 called Action emploi, modeled onSSP. This project continued until November 2005 and anevaluation is currently underway. Depending on theresults regarding the effectiveness of the program,consideration may be given to implementing a version ofit in the future.

In 2005 the Work Premium (prime au travail) wasimplemented; this program provides an incomesupplement to low-income workers in the form of arefundable income tax credit. Unlike APPORT, the WorkPremium is accessible to all low-income workers,whether they have children or not. Quarterly allocationscan be paid in advance to families with children whoface immediate financial needs. It is expected that thisprogram will be a better instrument to fight poverty andwill be more responsive to the needs of all low-incomeearners.

Don Norman, Program and Policy DevelopmentSpecialist with the Newfoundland and LabradorDepartment of Human Resources, Labour andEmployment, described a pilot program that offers anearnings supplement to single parents on welfare. Theprogram is delivered by using an external agency. Itbegins with six weeks of job-readiness training followedby four weeks of intensive job search. An evaluation ofthe program found that it was successful in reducingwelfare receipt and paid for itself in two and a half years.According to Norman, a number of key factorscontribute to the success of the program. First, clientswant to take part; they self-identify for the program and


5

3

are admitted if they meet the eligibility criteria. Second,the program is delivered by a community agency ratherthan directly by the department. Agency staff aresupportive of clients and play an advocacy role to makesure clients receive the benefits to which they areentitled. Third, participants need to have earnings toqualify for the income support component. The earningssupplement provides greater benefits to those with morehours worked and more earnings (up to a threshold level).Finally, the program provides ongoing employmentsupports. Norman said that the current plan is to expandthe program, which currently operates only in the greaterSt. John’s area, to the challenging area of westernNewfoundland, where unemployment is higher and thereare more rural areas.

Susan Williams, the Assistant Deputy Minister of thePeople and Skills Investments Division in the AlbertaDepartment of Human Resources and Employment,emphasized the provision of employment and trainingservices designed to ensure that low-income, low-skilledindividuals get the skills they need to advance in thelabour market. She said that Alberta currently spendsabout $288 million annually in federal and provincialfunds to help Albertans upgrade their labour force skills.Williams said that adopting an employment-focusedapproach was initially a huge shift for many of those inthe department. “We are not just about welfare anymore,”she said; now the first point of discussion is identifyingskill sets and removing barriers to work while discussionsabout income support come later. Staff spend more timeinteracting with employers, with training providers, andwith the broader community. However, she said that evenafter eight years the transition is still not complete. Somestaff have embraced the changes and are proud to cometo work and be able to talk to clients about a positivemenu of programs and services, she said. Other staff weremore reluctant to buy into the changes, and usuallymoved on to other areas of responsibility. Williamsindicated that she thought Alberta had been successful inputting in place an effective set of programs for helpingpeople participate in the labour market. The focus now ison integrating service delivery and simplifying programsso that they are easier for clients to understand.

Karen Glass, Director of Ontario Works, said that whenOntario Works was launched in 1996 it reduced benefitlevels, tightened social assistance eligibility requirements,and imposed an obligation on benefit recipients toparticipate in employment measures. During the secondhalf of the 1990s, there was a significant caseload

reduction, in part due to welfare reform and assisted by abooming economy. One result, Glass noted, has been thatan increasing proportion of the caseload is made up ofpeople with multiple barriers. In addition, the caseloadhas been slowly but steadily increasing since 2001.

Currently, Ontario Works offers both temporary financialassistance and employment supports to help people moveto paid employment. Benefit recipients must participatein activities that will support their progress towardsemployment. Glass said the benefit levels were “not therichest in the country, but they are not the lowest.”

Glass also stated that Ontario has also been attempting tosimplify its programs and procedures to make them morecost-effective. For example, a one-step applicationprocess has replaced a two-stage process. Glass spokeabout a number of recent changes that have been made tostrengthen incentives to work. These included asubstantial increase in the informal child-care allowance,enhancements to the earned income disregard thatexempts 50 per cent of all earnings from employment,and the provision of a transitional health benefit for thoseleaving social assistance for work.

Glass said “Our clients can get jobs, but they can’t keeptheir jobs.” Ontario’s JobsNow pilot project targetsrecipients who have at least 12 months of socialassistance receipt over a 36-month period. JobsNowparticipants are matched to jobs by a company hired byOntario Works that has experience in deliveringemployment placement and retention services.Participants receive ongoing job-retention supports for upto 18 months. A 42 per cent placement rate is anticipatedduring the pilot project, and payments to the companywill be based on the income assistance savings thataccrue as a result of clients being in jobs.

Glass described how the funding model for OntarioWorks delivery partners was also being changed to onethat was more results-based. Funding is being directlylinked to employment outcomes, such as earnings,increased employment, and enhanced employability.Some outcomes, such as increased employability, will bedifficult to measure. It will also be important to recognizethat external factors, such as how well the economy isperforming, can have a major influence on what happensto the caseload.

A summary of the proceedings of the Making Work PaySymposium is available on the SRDC Web site.

Spring 2006


6

Most evaluations of government programs focus onwhether a program improves, on average, the outcomesof persons who are offered the program. For instance,randomized experiments, such as the Self-SufficiencyProject (SSP), often focus on whether the averageoutcome of those in the program group is better than,same as, or worse than the average outcome of those inthe control group. SSP, for example, found that the offerof a generous earnings supplement to long-term welfarerecipients increased average earnings. However, thisfinding on its own may raise questions about how that

increase in average earnings is distributed acrossparticipants and across time. For example, was theaverage increase in earnings due to a broadlyexperienced increase occurring for most people at mosttimes? Or, in contrast, was the change narrowlyconcentrated at the top or the bottom of the earningsdistribution? Is an average increase in earningsconcealing decreases in some portions of the earningsdistribution and larger offsetting increases in otherportions? Have evaluators overlooked important, butoffsetting, effects of a program because the averageeffect is zero? Understanding the impacts of a programon the distributions of outcomes is important toproviding a complete evaluation of its effects.

In a recent SRDC working paper, Marianne P. Bitler,Jonah B. Gelbach, and Hilary W. Hoynes study theeffects of SSP on the distribution of earnings, transfers

(income assistance plus the earnings supplement forthose eligible to receive it), and total income (the sumof earnings and transfers). In doing so, the authors drawon rigorous techniques they developed for another papersoon to be published in the prestigious academic journalThe American Economic Review. The result of theirexamination of SSP is a series of easily understoodcharts that show the differences between the treatmentgroup and the control group distributions of earningsand other outcomes. These differences in distributionshave all of the beneficial statistical properties resultingfrom a randomized experiment.

Bitler, Gelbach, and Hoynes first calculate the earningsper month of every person in the program and controlgroups so that each participant has multipleobservations — one for each month of the study. Then,the authors order the monthly earnings observations ineach group from lowest to highest to obtain thedistribution of monthly earnings for both the programgroup and the control group.

The result is Figure 1 on the next page, which showsthe distribution of monthly earnings of the treatmentand control groups for Month 1 to Month 48 afterrandom assignment. Monthly earnings are on thevertical axis. The units on the horizontal axis areessentially percentages but, more precisely, are calledpercentiles or quantiles of the earnings distribution. Thebest known quantile is the 50th percentile or median,which, in this case, is the smallest level of earningssuch that 50 per cent of the observations have earningsbelow that level and 50 per cent have earnings abovethat level. Similar figures can be calculated for otherpercentiles. The solid grey line in Figure 1 presentsthese quantiles for the program group, while the dashedline presents the quantiles for the control group. Thefigure shows that most observations of monthlyearnings were equal to zero for the first quantiles on theleft-hand side of the chart. However, looking at 80thquantile on the horizontal axis and going up to the solidline shows that 80 per cent of the monthly earnings inthe program group were less than or equal to $784.Similarly, the dashed line shows that 80 per cent of the

SSP’s Impacts on the Distribution of Earnings

Understanding the impactsof a program on thedistributions of outcomesis important to providinga complete evaluationof its effects.


7

3

monthly earnings of the control group were less than orequal to $388.

Figure 1 shows that simply comparing the averagemonthly earnings of the program group ($334) with theaverage monthly earnings of the control group ($263)does not tell the whole story. In each group, earningswere below the level of average monthly earnings inmost months. For example, almost 80 per cent of thecontrol group monthly earnings were below the averageearnings of that group and 73 per cent of control groupmonthly earnings were zero. However, the highestmonthly earnings of both groups were substantiallygreater than the average monthly earnings.

Since the authors want to discover how SSP changedthe distribution of monthly earnings, they focus on thedifferences between the distributions of monthlyearnings in the program and control groups. In Figure 1,these differences are represented by the verticaldistance between the (solid) program group line and the(dashed) control group line. For the 80th quantile, thisdifference is equal to $396 ($784-$388). Similardifferences can be calculated for each point on thehorizontal axis. These differences are shown in Figure 2on the next page. One can see that the differencebetween the distributions of program and control groupmonthly earnings is $396 at the 80th quantile just as itwas in Figure 1.

Figure 1: Distribution of Monthly Earnings for Programand Control Groups, Months 1–48

Spring 2006


8

Figure 2 presents the main results of the analysis. Itshows that SSP had no effect on the low end of thedistribution of monthly earnings. Consequently, thedifferences between the two distributions on the lefthand side of Figure 2 are zero. However, at about the65th percentile, the differences between the programgroup and the control group distributions begin toincrease. They peak around the 75th percentile at justless than $470. The difference between the twodistributions then falls. At the very top of thedistributions, the differences appear to become negative,but these negative effects of SSP are not statisticallysignificant. At different points in the distribution, thesedifferences range from -$165 to $470, compared with anaverage treatment effect of $72. These results clearly

show that presenting the average treatment effect is notsufficient to characterize SSP’s effects on earnings.Bitler, Gelbach, and Hoynes present and discuss similarimpacts on the distributions of transfers and income intheir SRDC working paper.

The authors conclude that their work allows acomparison of the distributions of earnings and incomeunder two different policy regimes — in this case, theSSP program and control groups. While the methoddoes not, in general, reveal the impact of the policy onany specific person or group, it does reveal the impactof the policy on the distributions of earnings, transfers,and income, and these results provide important policylessons, the authors conclude. First, generous earnings

Figure 2: SSP Quantile Effects on the Distribution ofMonthly Earnings, Months 1–48


9

3

Spring 2006

supplements can increase earnings, but gains at the topand bottom of the earnings distribution are less likely toresult. Second, despite this generous incentive, thebottom parts of the earnings distributions are similaracross the program and control groups (since asignificant proportion of those in both groups continueto have little or no earnings). In fact, the largest gainsare concentrated in the top half of the distribution,though not at the highest end of the distribution, Bitler,Gelbach, and Hoynes note.

The results of their analysis have implications for allprogram evaluations, the authors conclude. They note

that it may be easy to dismiss a program that has littleor no effect on the average outcomes. However,programs whose average impact is zero may actuallyproduce substantial, and policy-relevant, effects on thedistributions of many important measures. Examiningdistributional impacts has the potential to uncoverwidely varying positive and negative effects that mightcancel each other out and be concealed if only averagesare considered.

Distributional Impacts of the Self-Sufficiency Project byMarianne P. Bitler, Jonah B. Gelbach, and Hilary W.Hoynes is currently available on the SRDC Web site.

Case Coordination and the Hard-to-Employ: An Initiative to Help the Long-TermUnemployed Move Into WorkThe Downtown Eastside (DTES) in Vancouver holdsthe title of one of Canada’s poorest postal codes. Thearea is synonymous with high rates of poverty,substance abuse, poor housing, and high rates ofunemployment. In a city with labour shortages and anunemployment rate of three per cent, in the DTES morethan one in five people are without work. While therehave been initiatives to try and resolve the complexsocial problems faced by individuals as well as by thearea, the statistics continue to demonstrate that thesocial problems persist.

In 2003 a two-year time limit on income assistancereceipt for individuals who are expected to work wasintroduced. This combined with the general consensusamong various stakeholders working in the DTES thatcurrent employment initiatives were inadequate toaddress the special employment needs of the “hard-to-employ” DTES residents. In response to these concerns,the Vancouver Agreement implemented a three-yearinitiative in the fall of 2004 to help long-term

unemployed residents of the DTES to enter work byproviding them with a one-to-one case coordinationservice. To do this, the Vancouver Agreement received agrant of $3.25 million from the Ministry ofEmployment and Income Assistance (MEIA) todevelop, implement, and evaluate the CaseCoordination Project (CCP). SRDC is conducting anevaluation of the program to determine if, and how,providing intensive one-to-one support to long-termunemployed people helps them move into work.

What is case coordination?SRDC conducted a review of 10 programs that includedintensive case management or coordination approachesfor the hard-to-employ. It found that the programsshared a range of 13 components, listed in the table onthe next page. The most common features were

l financial or other aid to reduce barriers toassistance,

l intensive personalized assistance,


10

Components of Case Coordination

Component Present in the CCP Access to Other Services

Counselling ü

Customized action plans ü

Education and training ü Make referrals

Financial or other aid to reduce barriers to assistance ü

Health services ü Make referrals

Intensive personalized assistance ü

Individually tailored resources ü

Job-preparation and job-search assistance ü

Life skills development ü

Service referrals ü

Site level interagency coordination ü

Vocational training ü Make referrals

Whole client needs assessment ü


11

3

Spring 2006

l individually tailored resources,

l job-preparation and job-search assistance, and

l whole client needs assessment.

The evidence suggests that programs that are designedto address a broad range of employment barriers maybe more effective in helping the hard-to-employ movetowards work. The design of the CCP originated fromthe practical experience of practitioners rather than anysingle model, but it incorporates the majority of thecomponents listed in the table and highlights afundamental principle of the project, namely to tacklethe underlying causes of the barriers to employment.Currently, there is very little known about whichcomponents or combination of these are most importantin case coordination.

Case coordination in the DTESThe case coordination model being implemented is adecentralized one in which six case coordinators areemployed by six service agencies. The service agenciesare responsible for managing the individual casecoordinators, and the Vancouver Agreement hasresponsibility for them in terms of how they implementthe model.

To be eligible for CCP, clients have to be on incomeassistance and residing within the catchment area oftwo local MEIA offices. MEIA staff determine whethera client is eligible according to a set of agreed criteria,and the case coordinators decide if a client is suitablefor the project.

Once a client has been accepted into the project, thecase coordinator completes an assessment and developsan individualized employment plan. The casecoordinator then works with the client to achieveagreed goals that will help him or her move towardwork. The case coordinators have access to clientintervention funds to help them achieve these goals.This individual and local pot of money means that thecase coordinators are able to respond quickly andcreatively to client needs. To facilitate the sharing ofpractice information, the case coordinators meet bi-weekly to case conference as well as to discuss the

types of interventions and strategies they areimplementing.

Evaluating the Case CoordinationProjectThe goal of this evaluation is to answer the centralquestion: Does providing intensive one-to-one supportservices and guiding individuals back to worksuccessfully reconnect them to the labour market, andhow is this achieved?

The specific research questions are as follows:

l What does it take to successfully implement theintervention — what is case coordination withinthis context?

l How do hard-to-employ individuals move towardswork — what processes are involved?

l Does the project lead to less dependence on incomeassistance beyond what would have happened in theabsence of the project?

A comprehensive research design is being used toaddress these questions. Implementation research willbe used to explore issues around practice and tounderstand the processes involved. A non-experimentalimpact analysis will be conducted to explore the effectsof the project in terms of receipt of income assistanceas well as the amount received and will rely onadministrative data from MEIA. In addition, casestudies will be conducted with clients to explore theactual process as well as the experience of casecoordination. The case studies will focus on themovement of clients towards work and explore theinterventions that help or hinder this.

The clientsThe clients referred to the CCP are described as beingvery “wounded and needy.” They lead very fracturedand chaotic lives. Almost all have tried otheremployment initiatives but have been unable to sustainemployment and therefore cycle between welfare andwork.


12

During the first year, almost 200 clients have beenreferred to the project. Three quarters are men, most ofwhom live alone. Over half of the clients are less than45 years of age. The majority of the clients have activeaddictions. More than half the clients experiencemultiple barriers including health and addictionproblems, issues from their personal history, and lack ofadequate housing. Based on these statistics, the projectappears to be enrolling clients who fit the targetpopulation for the intervention.

The case coordinators have spent a large proportion oftheir time and efforts on making the initial connectionwith clients, building trust, and addressing clients’ basicneeds. Many of the clients have a very checkeredhistory with intervention programs and are suspiciousabout the project (especially new initiatives) andwhether it is any different than others they haveattended. Therefore, it is important that the casecoordinators establish trust with their clients; otherwisethey will not be able to tackle some of the underlyingproblems, which perpetuate the “welfare–work–welfare” cycle.

To do this, some meetings between clients and the casecoordinators take place in their offices, others involvegoing for lunch or coffee. Many of the case coordinatorsuse food as an incentive to encourage clients to show upfor these initial meetings. Findings from focus groupsconducted with clients suggest that this approach iseffective. Clients reported that getting something to eatwas the main reason to turn up in the beginning. It wasonly as the clients recognized that the project wasdifferent that the meeting itself was seen as a reason toattend.

The client assessments conducted by the casecoordinators reveal that the majority of clients havesubstantial basic needs that have to be addressed beforetackling issues related to education, training, oremployment. Basic needs include health, housing,personal hygiene, adequate food, and clothing. For someof the clients it is simply a matter of being referred onto another service agency that can provide theserequired services and supports.

However, the majority of clients require considerablesupport to complete tasks such as getting anidentification card, attending a health appointment, orfinding housing. This often involved helping clientscomplete forms and photocopy documents andaccompanying them to the appointment or office. Whilethese tasks were often the gateway to other services,which is significant in its own right, equally importantwas the initiation of a process of moving forward andconvincing the client that it was possible to addressproblems.

Providing clients with an opportunity to succeed is seento be important as well. Many of the clients whoattended the focus groups had accepted the label of“failure.” They had failed so many times in the past thatthey simply stopped trying. They saw no point inattempting to change their situation or circumstances,because experience had taught them that they would notsucceed. Failure often triggered a downward spiral, sodepression, mental health problems, and addictions alldeteriorated. The role of the case coordinators indemonstrating that there is potential for change issignificant for those clients ready to engage with theproject.

The next stepsSRDC’s evaluation study is designed to run alongsidethe project. The first year of the evaluation concentratedon understanding how the project operated, includingprocesses and practices. The second year of theevaluation will examine in detail the actual process ofcase management so that it will be able to explain whycertain interventions with clients are successful andothers are not. Interviews with the case coordinators,service providers, and stakeholders will be repeated, butthe emphasis will be on changes within the projectstructure, organization, procedures, and practices. Asecond round of focus groups will be conducted withclients to gain greater insights into the impact that casecoordination has upon the lives of the clients. Finally,work on the non-experimental analysis will begin,which will explore the effect of the project on incomeassistance receipt.


13

3

Spring 2006

Institutional Factors Can AffectPolicy ReformsLarge changes in social policies and programs can havelarge but uncertain effects on social norms and on theexpectations both of the staff who administer programsand of the clients who benefit from them. Thesechanges can potentially have large positive or negativeeffects on clients that are in addition to the directprogram impacts that are usually measured in theevaluation of social programs. However, the changes tosocial norms and expectations and their effects onclients and staff are difficult to understand, predict, andmeasure.

Robert Moffitt, the Krieger-Eisenhower Professor ofEconomics at Johns Hopkins University, explained tothe recent Making Work Pay Symposium (see p. 1) thatonce a program begins to affect a widespread change inbehaviour, social norms and expectations begin tochange as well, and sometimes these changes are thecentral goal of the program. Moffitt gave as an examplethe US welfare reforms in the 1990s, which, amongother things, placed limits on how long clients couldreceive welfare benefits. Moffitt said a key programgoal was to change the “culture of welfare” and theexpectations of those receiving welfare benefits. Thegoal was to create an expectation that welfare receiptwas temporary and that when people were on welfare,they were expected to work. “Whether you like thereforms or you don’t, I think the reforms achieved that,”Moffitt said.

He also described “institutional effects” as whathappens when an innovative program is integrated intoexisting administrative structures, especially if theprogram runs counter to existing administrative-culturalnorms. For example, prior to the most recent welfarereforms in the US, the welfare system had the culture ofa cheque-writing organization; most of the effort wentinto establishing eligibility and ensuring that everyonereceived the benefits to which they were entitled. These

institutional factors can impede the promotion ofpolicies to encourage work. Moffitt gave the example ofearnings disregards — which encourage work byallowing welfare clients to keep part of their earnings— which were widely thought to function poorly in theUnited States. For a working client, case workers haveto do a lot of work — checking earnings, getting

information on work-related costs — that is notrequired in the case of a non-working client. As a result,he suggested, some caseworkers regard having workingpeople on the welfare rolls to be a workload burden andthey do not actively encourage clients to take advantageof these provisions. A goal of US welfare reforms wasto change the cheque-writing culture into one in whichthe attainment of work was a major organizationalobjective.

Large changes in socialpolicies and programs canpotentially have largepositive or negative effectson clients that are inaddition to the directprogram impacts that are usually measured in the evaluation of social programs.


14

Moffitt also pointed out that it may be possible to run anew, exciting, small-scale experiment with simplifiedadministration and eligibility conditions and a coregroup of dedicated staff who want to make the projectwork. However, these conditions may not be replicablewithin a large administrative structure and,consequently, the impacts seen in an experiment maynot be observed after full implementation. He said theeffects of these social and institutional factors aredifficult to achieve, model, and predict. For example,small-scale pilot projects often used in randomassignment experiments are not large enough togenerate social and institutional effects.

Several provincial government speakers also told theMaking Work Pay Symposium about their experienceswith social and institutional change. Not all of thesechanges were improvements. Rick August, ExecutiveDirector of Strategic Policy in the SaskatchewanDepartment of Community Resources and Employment,told the symposium that, at times, traditional incomesupport programs undermined the traditions of work insome communities. August recounted how HaroldLarade from Saskatchewan’s Cowessess Reservedescribed the coming of the welfare system to hiscommunity when John Diefenbaker was prime minister.“I was 18 when they started handing out welfare money.I knew welfare was going to spoil us. Even the able-bodied quit working. If you worked, you could not getwelfare. Some men never worked again,” Augustquoted Larade as saying.

August said Saskatchewan has developed a distinctunderclass of individuals with little education,inadequate literacy or numeracy skills, and little or no“personal work culture.” He said more recent changeshave attempted to make work more central to the socialcontract. “Unfortunately, undoing a half a century ofmarginalizing social policy does not happen overnight.It is not a simple task,” he said.

Saskatchewan Senior Policy Analyst Doug Scott alsodescribed how paternalistic and intrusive policies of hisdepartment can unintentionally foster dependencyamong clients. “Clients would often describe their livesas awful before they got case worker ‘X’ and thenthings got wonderful because case worker ‘X’ took careof them. That is a pretty paternalistic relationship,” hesaid. Or, the department would undertake to pay bills on

behalf of clients, for example paying a client’s utilitiesbills directly to the utility companies in order to preventthe disconnection of services. While this solved theimmediate problem, it left the client with no experienceof budgeting for utilities when they left welfare. Thedepartment had a similar arrangement for rentpayments. Scott said the department is slowly extractingitself from these types of arrangements. “You can’tteach people to behave responsibly by takingresponsibility away from them,” Scott said.

Scott said departmental culture is changing as well.Previously, the service structure was directed primarilytoward benefit administration with well over 150 casesper case worker. There was not enough time for workersto do much other than administer benefits; there werevery limited case planning activities despite a formalrequirement to do so. Caseworkers had little contactwith clients except for a mandatory annual review.More recently, however, much of the benefitadministration has been shifted to contact centres,leaving more time for department officials to do caseplanning and employment counselling. Enthusiasm forthe change was initially high but has since waned forsome staff, Scott said. Some department personnel havesuccessfully made the adjustment to become career andemployment counsellors but others have not, Scott said.

Susan Williams, the Assistant Deputy Minister of thePeople and Skills Investments Division, Alberta HumanResources and Employment, said that her department isstill in the midst of a cultural change, begun in 1997,that places far greater emphasis on training and skillsupgrading, with a lower emphasis on income support.Williams said that adopting an employment-focusedapproach has been a huge shift for many of those in thedepartment, and even after eight years the transition isstill not complete. Staff face a greater need to interactwith employers, training providers, and the broadercommunity, Williams said. Some of them haveembraced the change, while others have moved on toother areas of responsibility.

The study of cultural and institutional change hasattracted interest in economics but is a well-developedarea of study in fields such as sociology.


15

3

Spring 2006

SSP, Education, and Welfare ExitsIncreased education and labour force skills developmentare often held to be an important method of improvingthe labour market performance of low-incomeCanadians. SRDC has just published two workingpapers — one by Chris Riddell and W. Craig Riddelland another by Jorgen Hansen — that shed some lighton the educational upgrading of single parents onwelfare. In some aspects, the authors of these papersshare similar views; in other respects they come todifferent conclusions.

The Self-Sufficiency Project (SSP) offered generousearnings supplements to single parent, long-termwelfare recipients if they found full-time work quickly.Riddell and Riddell studied educational upgradingamong the long-term welfare recipients in the SSPRecipient study in British Columbia and NewBrunswick. Hansen studied the effects of educationamong participants in the SSP Applicant study in whichnew applicants for income assistance benefits in BritishColumbia became potentially eligible for the SSP’searnings supplements if they remained on welfare forone year.

Both Riddell and Riddell and Hansen conclude that asubstantial amount of educational upgrading is beingundertaken by SSP participants. At the start of theRecipient study, Riddell and Riddell find that more thanone half of SSP Recipient participants had notcompleted secondary school; however, among thosewithout a high school diploma, about 20 per cent hadcompleted secondary school 54 months later. During thesame period, more than one fifth of those who had neverbeen enrolled in a community college or trade schoolprogram at the start of the study had done so by the endof the study.

Similarly, Hansen finds that in the Applicant study35 per cent of those with less than a high schooldiploma at the time of the initial interview upgradedtheir education during the following six years. About45 per cent of the Applicant study sample used byHansen had less than high school at the start of thestudy, but only 28 per cent had less than high school sixyears later — a decrease of 17 percentage points.Among high school graduates, educational upgrading

was also common. Seventeen per cent had completedvocational school at the start of the study compared withtwenty-three per cent six years later, Hansen found.

Both sets of authors note that, in light of the substantialamount of educational upgrading that already occursamong welfare recipients, introducing new educationalprograms for those receiving income assistance wouldprovide services to a significant number of people whowould have upgraded their education and job skillswithout such a program and, as a consequence, theeducation program would produce much smallerimpacts than would be suggested by a simplecomparison of participant education levels before andafter the introduction of the program. Hansen also notedthat research studies that incorrectly assume educationlevels remain relatively constant over time may havebiased results for variables that are correlated witheducational upgrading.

Riddell and Riddell find that the offer of the earningssupplement reduces the educational upgrading of theprogram group relative to that of the control group. Theauthors suggest that it is possible the financial incentiveencouraged program members to find full-timeemployment quickly with the consequence that they hadless time to upgrade their education. In support of thishypothesis, the authors note that the difference betweenthe educational courses taken by the program group andthe control group is greatest when the employment gapbetween the two groups is largest. The difference ineducational upgrading between the two groups is largestat the secondary school level but is also observedamong community college programs. The authorsconclude that, overall, SSP had the adverse side effectof reducing participants’ participation in education.

Riddell and Riddell “match” participants who upgradedtheir education during the study with participants whodid not. They find that that “upgraders” generallyachieve larger gains in employment and wage rates thantheir matched counterparts who did not upgrade theireducation. For example, the authors find that highschool completion is associated with an employmentgain of 13 percentage points. Both secondary schoolcompletion and enrolment in a college or trade school


16

SRDC is pleased to announce that it has been selectedby Human Resources and Social Development Canada(formerly Social Development Canada) to manage andevaluate a new pilot project that will deliver enrichedchild-care services to francophone pre-schoolers inminority francophone communities. This project is partof the Action Plan for Official Languages that wasoriginally announced by the federal government in March 2003.

The Child Care Pilot Project is designed to assist indeveloping the French language skills, the francophoneidentity, and the learning capacity of children who live inminority francophone communities. The services beingoffered to the children and their families will have twomain components. The first component will take placedirectly in daycare centres and will offer preschool-agedchildren a range of activities intended to develop theirFrench language abilities and their sense of identity. Thesecond component will involve the participation ofparents and will consist of a series of literacy workshopsfor children and their parents with the main purpose ofawakening children’s interest in reading.

SRDC’s role will be to coordinate the implementation ofthe program and then to rigorously evaluate it to assessthe extent to which it achieves the desired outcomes.Five minority francophone communities across Canadawill be selected to be the test sites for the project and,within each community, 40 three-year-olds will beselected to take part. The project will use a randomassignment design; half of the children will be assignedto the program group; the remainder will be assigned toa control group that will not be eligible to receive theenriched services. Data will be collected on a regularbasis from all the children and from their parents inorder to determine the impacts of the enriched serviceson children’s development.

For this project, SRDC will be working in collaborationwith Human Resources and Social Development Canadaas well as with a research group led by Dr. J. DouglasWillms, KSI Research International, which is responsiblefor developing the data collection instruments. The valueof the contract is $2.8 million over the next two years.

SRDC Wins New Contract for aFrancophone Early Childhood Project

program is associated with an 18 per cent wage gain.After a series of statistical tests, Riddell and Riddellconclude that the positive wage and employment gainsexperienced by dropouts who later completedsecondary school, and the positive wage gainsexperienced by college and trade school enrollees, area consequence of the educational investments madeduring the study period.

Riddell and Riddell also find that upgraders are lesslikely to remain on welfare. However, this may be dueto unobserved differences between upgraders and non-upgraders rather than being attributable to theiradditional education. Overall, the authors conclude thatinvestments in formal education — such as completingsecondary school — can yield significant benefits forsingle parents on welfare.

Hansen focuses on the effects of education on welfareexit and re-entry. He says that it is conventionallyassumed that increased formal education results inreduced welfare usage, and, consequently, programs to

increase the formal educational levels of currentwelfare recipients will lead them to leave welfareearlier and return to welfare less frequently. Hansenstates that this conventional wisdom is incorrect, andhe concludes that more formal education does notcause lower welfare use. Instead, favourableunobserved characteristics — such as labour-marketability, motivation, and preferences — cause somewelfare clients to have both higher levels of educationand shorter, less frequent spells on welfare.Consequently, Hansen concludes that policies aimed atimproving the educational attainments of formerwelfare recipients may have only limited effects on thereliance on income assistance.

The working papers Educational Upgrading and ItsConsequences Among Welfare Recipients: EmpiricalEvidence From the Self-Sufficiency Project by ChrisRiddell and W. Craig Riddell and The Effects ofHuman Capital and Earnings Supplements on IncomeAssistance Dependence in Canada by Jorgen Hansenare available on the SRDC Web site.


17

Spring 2006

Bulletin Board

Publications

Distributional Impacts of the Self-Sufficiency Project by Marianne P. Bitler, Jonah B. Gelbach,and Hilary W. HoynesThis working paper examines the impacts of the Self-Sufficiency Project (SSP) on the distributionof earnings, transfers, and income using quantile treatment effect (QTE) estimation. QTE estimatesshow that the SSP program had heterogeneous impacts on the distributions of earnings, transfers,and total income — heterogeneity that would be missed by looking only at average treatmenteffects. The authors argue that these findings are consistent with labour supply theory, in whichworkers respond to financial incentives by changing their hours worked and, in some cases,reducing the reservation wages at which they will just be willing to take a job.

The Value of Non-market Time Lost During the Self-Sufficiency Project by David H. Greenbergand Philip K. Robins This working paper re-examines the original cost-benefit analysis of Self-Sufficiency Project(SSP). It looks at the non-market time that program participants lost by increasing their work effort,and it considers whether different conclusions regarding the success of the program might bereached if SSP’s effect on non-market time is taken into account.

Understanding the Dynamic Effects of the Self-Sufficiency Project Applicant Study by David Card and Dean R. HyslopThis working paper focuses on the impact of the Self-Sufficiency Project (SSP) Applicant study onIA participation. It shows that the Applicant study created three incentives: (1) an eligibilityincentive for everyone in the program group to remain on welfare for a year to become eligible forthe subsidy, (2) an establishment incentive for members of the treatment group who satisfied thewaiting period requirement to find a job and leave welfare within the next 12 months, and (3) anentitlement incentive for those who established SSP eligibility to work full time and remain offwelfare in the three-year period during which subsidy payments were available. Nearly all of thedelayed exiters — people who delayed their exit from IA in response to the supplement — left IAwithin two to three months of the end of the waiting period and became entitled for the SSPsubsidy. The paper estimates that roughly two thirds of the peak impact of SSP on welfareparticipation was attributable to the time-limited entitlement feature of SSP. Those who receivedsubsidy payments tended to remain off welfare, even after the subsidy payments ended. About 60 to90 per cent of the entitlement incentive effects experienced by members of the treatment groupduring their three-year period of subsidy eligibility persisted in the immediate post-entitlementperiod, though the effect faded relatively quickly.

Evaluating Search and Matching Models Using Experimental Data by Jeremy Lise, Shannon Seitz, and Jeffrey SmithThis working paper introduces an innovative test of the quantitative performance of search andmatching models that uses information becoming available through social experiments. It presentsa prototypical Pissarides matching model and calibrates it to control group data from the Self-


18

Sufficiency Project (SSP). Then a program group is simulated from a randomized experimentwithin the model, and the outcomes are compared with those of the program group from SSPto determine whether the model is an appropriate tool to use in estimating the behaviouralresponse to SSP.

The Effects of Human Capital and Earnings Supplements on Income AssistanceDependence in Canada by Jorgen HansenThis working paper uses data from the Self-Sufficiency Project (SSP) Applicant study toanalyze the effects of educational attainment, work experience, work-related training, andearnings supplements on income assistance (IA) dependence. The main objective of theempirical analysis is to estimate the effects of these factors on both IA exit and IA re-entryrates. This paper suggests that, contrary to conventional wisdom, formal education has nosignificant causal effect on IA use. The paper also finds substantial educational upgrading bysingle parents in receipt of IA. Higher formal education does not cause lower welfare use.Instead, the paper finds that favourable unobserved characteristics — such as labour marketability, motivation, and preferences — cause some welfare clients to have both higher levelsof education and shorter, less frequent spells on welfare. Consequently, the paper concludesthat a policy aimed at improving the educational attainments of former welfare recipients mayhave only limited effects on the reliance on income assistance.

An Econometric Analysis of the Incremental Impact of SSP Plus by Jeffrey Zabel, Saul Schwartz, and Stephen DonaldThis working paper evaluates both the short-term and the long-term incremental impact ofSSP Plus on the duration of periods of full-time employment and periods during whichparticipants were not working full time. It examines whether SSP Plus had any additionalimpact on these durations beyond those created by the SSP earnings supplement alone.Estimates of the “effect of the treatment on the treated” are generated, where the “treated” aredefined as those in the SSP Plus and regular SSP program groups who qualified for theearnings supplement by finding a full-time job within the first 12 months after randomassignment. The study finds evidence of the impact of SSP Plus on unemployment andemployment durations during the period when those who took up the earnings supplementwere still eligible for that supplement.

An Analysis of the Impact of SSP on Wages by Jeffrey Zabel, Saul Schwartz, and Stephen DonaldThis working paper examines wage progression as one theoretical explanation for theexistence of a long-term impact on employment due to the Self-Sufficiency Project (SSP). Inorder to qualify for the supplement, program group members had to find full-time work withinone year of random assignment and, once they qualified, they received the supplement whilethey were working during the subsequent three years. Because of this additional incentive towork, these “take-up” group members acquired more full-time work experience thancomparable control group members. The greater full-time work experience should imply, onaverage, that at the end of the follow-up period the wages of take-up group members werehigher than that of comparable control group members. This is referred to as “relative wageprogression.” Because of the higher wages, take-up group members should be more likely to

19

3


work than comparable control group members even after the supplement period has ended.That is, relative wage progression is one avenue through which SSP might have had long-termimpacts on the ability of recipients to work full time.

Educational Upgrading and Its Consequences Among Welfare Recipients: EmpiricalEvidence From the Self-Sufficiency Project by Chris Riddell and W. Craig RiddellThis working paper examines the extent and nature of educational upgrading amongparticipants in the Self-Sufficiency Project (SSP) and the effect that increases in educationalattainment had on the employment, earnings, and reliance on welfare among these long-termwelfare recipients. The study shows that there is a substantial amount of educational upgradingamong single parents on welfare. SSP program group members acquired less additionaleducation than their counterparts in the control group. One possible explanation is that SSPencouraged full-time employment, leaving less time for acquiring additional education. Thosewho upgraded their education generally achieved larger gains in employment and wage ratesthan did their counterparts who did not acquire additional education, the study concludes.

The Effect of the Self-Sufficiency Project on Children by Piotr Wilk, Michael H. Boyle,Martin D. Dooley, and Ellen L. LipmanThis working paper assesses whether the Self-Sufficiency Project (SSP) affected the health,behaviour, and academic achievement of children and whether these effects were sustainedbeyond the period of the intervention. This study assesses the overall effect of SSP on all thechildren in the program group as well as the effect on those children whose parents received atleast one supplement payment. The results suggest that if a program like SSP was to beintroduced to the general population, it would be unlikely to affect children’s health, behaviour,and academic achievement, regardless of the level of program take-up.

Human Capital and Search Behaviour by Audra Bowlus, Lance Lochner, Chris Robinson,and Yahong ZhangThis working paper uses a structural search model to examine how changes to differentparameters in the Self-Sufficiency Project Applicant study might have changed the outcomes ofthe study. The model allows for the simulation of changes in the length of time an individual isrequired to stay on income assistance, the length of time within which a full-time job has to befound, the length of time the bonus can be received, and alternative levels of generosity of thebonus.

Making Work Pay (summary of the “Making Work Pay” symposium)This publication is a summary of the proceedings of the Making Work Pay Symposium, whichwas held in Ottawa, Canada, on November 15 and 16, 2005. Senior federal and provincialgovernment representatives, international experts, and leading academic researchers who met atthis symposium heard both research findings and practical experience about the latest efforts toencourage work among welfare recipients. In addition, several leading academic researcherspresented research that used data from Canada’s most famous social policy experiment — theSelf-Sufficiency Project.

The Making Work Pay Symposium was funded by Human Resources and Skills DevelopmentCanada (HRSDC), and was organized by SRDC.

Spring 2006


20

Events

SRDC participates in a workshop on social experimentationOn March 27 and 28, 2006, the Policy Research and Coordination Directorate of HumanResources and Social Development Canada (HRSDC) convened a workshop on socialexperimentation. A broad cross-section of HRSDC staff attended, as did representatives fromService Canada and a number of provincial governments. The lead-off speaker was JeffSmith from the University of Michigan who explained the benefits of a random assignmentdesign for estimating program impacts. SRDC’s Executive Director, John Greenwood, andSheila Currie, Director of Project Operations, spoke about a number of operational issuesassociated with designing and implementing social experiments. The workshop also heardfrom Arthur Sweetman from the Queen’s University School of Policy Studies and AndrewParkin from the Canada Millennium Scholarship Foundation as well as from tworepresentatives of American organizations that have been active in conducting social policyexperiments — Peter Schochet from Mathematica Policy Research and James Kemple fromMDRC. During the course of the workshop, time was set aside to explore how randomizedexperiments could be applied to specific issues of interest to those in attendance.

Survey of Parents of Kindergarten Students in Manitoba

The Healthy Child Manitoba Office (HCMO) has commissioned SRDC to conduct atelephone survey of parents of kindergarten students in Manitoba later this year. It will be thesecond time SRDC has conducted this successful survey. HCMO shared results from the firstsurvey, completed in 2004, with professionals on the front line, school divisions, and parentsin order to improve outcomes for children and families.

SRDC with help from POLLARA Inc., the largest Canadian public opinion and marketingresearch firm, will contact parents in 36 school divisions in May and June 2006. The EarlyDevelopment Instrument (EDI) developed by researchers at the Offord Centre for ChildStudies, McMaster University, is administered by teachers of kindergarten students inManitoba. The EDI results will help Manitoba to know how well its children are prepared tobegin their school lives ready to learn and to enjoy the experience. The telephone survey willcollect detailed information from parents about the experiences of the children and theirfamilies. Parents will be asked for their permission to allow SRDC to combine their child’sEDI results with the survey information, which will allow the analysis to reveal more aboutthe relationship between current kindergarten students’ readiness to learn and their familybackground and experiences.

SRDC looks forward to being part of this collaborative process to strengthen the use ofresearch evidence to improve outcomes for children and families.

table of contents new ceo for srdc · newsletter more attractive and easier for you to read, it has...

Documents