50 %
50 %
Information about Freeman_Your_Job_Love_it_or_Hate_it

Published on November 28, 2008

Author: aSGuest4337


Your Job: Love it or Hate it?The economics of job satisfaction : Your Job: Love it or Hate it?The economics of job satisfaction Richard B. Freeman Harvard University, NBER Centre for Economic Performance, LSE Aberdeen, June 29, 2007 Take this job and shove it : Take this job and shove it I ain't workin' here no more.My woman done left,An' took all the reasons I was workin' for.You better not to try to stand in my way,As I'm a walkin' out the door.Take this job and shove it,I ain't workin' here no more I've been workin' in this fact'ry,For nigh on fifteen years.All this time, I watched my woman,Drownin' in a pool of tears.An I've seen a lot of my good folk die,Had a lot of bills to pay.I'd give the shirt right off of my back,If I had the guts to say Well, that foreman, he's a regular dog, The line boss is a fool.Got a brand new flat-top haircut; Lord, he thinks he's cool.One of these days, I'm gonna blow my top,And sucker, he's gonna pay. Lord, I can't wait to see their faces,When I get the nerve to say: Take this job and shove it The Take this Job and Shove it Model : The Take this Job and Shove it Model Working for family Permanent tiresome work Low wages Nasty unintelligent bosses Response: Quit Slide 4: The Sunday Times 100 Best in UK 1. W L Gore & Associates 2: Sandwell Community Caring Trust 3: Pannone 4: Beaverbrooks the Jewellers 5: Data Connection The Fortune 100 Best Companies in US 1 Genentech 2 Wegmans Food Markets 3 Valero Energy 4 Griffin Hospital 5. W.L. Gore & Associates The Great Job model : The Great Job model Highly profitable – protected by intellectual property, non-profit Create family/team environment Often shared capitalist modes of pay + employment involvement in decisions Very careful recruiting Non-hierarchical open door management Low turnover  “voice” channels for dealing with problems This Talk : This Talk 1. Why job sat is important 2. Measurement issues 3. Workplace effect 4. Does changing job raise satisfaction? 5. Consequences of job sat – turnover, productivity, non-work outcomes Framework Model of Job Satisfaction : Framework Model of Job Satisfaction Xw  Job Satisfaction  Work behavior (quit/form union/absences); personal behavior (health); externality on others (SOs, neighbors, go postal) Xp  Non Work satisfaction Personal behavior (Health); work behavior ; externality on others (SOs behavior) Job Satisfaction Non work satisfaction Job Satisfaction//Non work satisfaction  Life satisfaction Problem of work-life balance -- more time/effort to job sat other forms of sat? Two conceptual problems – satisfaction = f( Actual vs Expected); Expected = g (Actual) -- adaption Correlation vs Causality: four hypotheses : Correlation vs Causality: four hypotheses H1: “It’s all you” – heritability and omitted personal factor. Direct estimates of heritability of job sat using twins/others; longitudinal same person H2: “It’s workplace” – do attitudes/behavior of persons have large workplace component? Do changed workplace alter job satisfaction? H3: “It’s the match” – comparable to dating/ marriage H4: “Not too many arrows” -- Xw, Xp work through satisfaction variables so that Xw Job Satisfaction  Nonwork satisfaction  Non work behavior Slide 9: The Genetic Hypothesis Why Care?1. Job satisfaction is part of life satisfaction : Why Care?1. Job satisfaction is part of life satisfaction Analysis of ISSP surveys for 2002 (Family and Changing Gender Roles), which asked about satisfaction with life; with family and job, showed that family satisfaction matters most, but job matters a lot, as well. The 2003-2006 Aberdeen Epicurus EU study found that “Job satisfaction secret of happiness” (The Scotsman, June 30 2006) Helliwall and Huang found that job satisfaction, how other time spent, health, about equally important (Canadian survey, NBER wp 11807, 2005) Slide 11: 2. Your partner/family cares because you bring your job satisfaction or misery home. Work-life balance No experimental evidence; can look at satisfaction and future break-ups in longitudinal surveys Positive correlation between job satisfaction and family satisfaction that has not been unpacked could be unobserved person could be family   job 3. Your colleagues care because you can make their life pleasant/miserable “Going postal” – laid off or unhappy worker gets gun and shoots up workplace, regular event in US CNN NEWS Tuesday, January 31, 2006 Seven dead in California postal shooting Shooter turned gun on herself, authorities say Slide 12: 4. Employer cares because job satisfaction  profits lower quits  reduced turnover costs more cooperative/peer monitoring lowers supervisor costs accept lower pay – compensating differential more productive in production function sense 5. Social scientists care because can help predict behaviour and test theories of workplace quit rates & absenteeism & productivity desire for unions/other forms of representation non-work behavior theories of group activity, social capital principal agent models;incentive models Dicey to economists bcs it is subjective : Dicey to economists bcs it is subjective Subjective variable involves interpersonal comparisons. My ecstasy is your eh; also creates issues of adjustment – your experience or your expectations affects what you report: the puzzles of women, education, union; differential tradeoffs for temporary and permanent workers But job satisfaction is connected to behaviour. It predicts some future actions better than objective variables. Analogy is with doctor asking “how are you feeling”? – more important than some “objective” tests Biological markers: Saliva – measures of stress; Brain scans Market test: A 1996 review article estimated that by 1991 more than 12,400 studies had been published on job satisfaction so probably 15,000-20,000 now. 2. Measuring Job Satisfaction : 2. Measuring Job Satisfaction Single question: How would you rate your satisfaction with your current job on a scale of 1 to 5 (5 being highest)? Most people report in 4-5 highly or somewhat satisfied groups. But distribution varies across surveys and time and countries. Lowest in East European transition economies. Reliability (Test/Retest) and Stability : Reliability (Test/Retest) and Stability Test-retest reliability measures of job satisfaction on the same job for satisfaction questionnaires are on the order of 0.80 (van Sanne, Sluiter, Verbeek, Frings-Dresen, 2003, table 3) In the NLSY, the correlation between the job satisfaction of workers who stay on the same job in two periods ranges from 0.35 to 0.45. Other studies of the attitudes of workers at two periods of time suggest that correlations of attitudes are on the order of 0.50 to 0.70[1]. We don’t know why the job satisfaction changes in the same job over short periods. It could be a change in job conditions … a change in expectations … or a change in some outside factor. [1] Rode (2004) estimates a correlation of 0.49 for job satisfaction for persons in the first and second waves of the Americans’ Changing Lives survey, where the waves were three years apart; Bowling, Beehr and Lepisto (2006) report a correlation of 0.53 on job satisfaction for respondents in the Adult Longitudinal Panel report correlations ranging from 0.58 to 0.68 for measures of organizational commitment, job involvement, career commitment and career satisfaction. Cote and Morgan (2002) report a correlation of 0.48 for a sample of 111 workers at two points of time separated by four weeks. US surveys show different trends, 1995-2005 : US surveys show different trends, 1995-2005 Conference Board study “U.S. Job Satisfaction Keeps Falling” (2005) “Americans are growing increasingly unhappy with their jobs, The Conference Board reports today. The decline in job satisfaction is widespread among workers of all ages and across all income brackets. Half of all Americans today say they are satisfied with their jobs, down from nearly 60 percent in 1995.” National Opinion Research Corporation yearly job satisfaction question on general social survey found NO CHANGE OVER SAME PERIOD, as does Gallup Poll Multiple Question Detailed employee opinion surveys : Multiple Question Detailed employee opinion surveys Examples: academics -- Minnesota Satisfaction Questionnaire; business -- Mayflower Group of up to 50 US firms) How satisfied are you:...that your supervisor keeps you well informed about what's going on? ...that your views and participation are valued?...with the morale of the people with whom you work?...that your work gives you a feeling of personal accomplishment? ...that you receive appropriate recognition for your contributions?...with the empowerment you have to influence the quality of your work?...with the reasonableness of your work responsibilities?...with your ability to balance family life and work life?...that your compensation matches your responsibilities?...with your job security?...with the amount and frequency of informal praise and appreciation you receive from your supervisor? UK Amicus Union Short Questionnaire: Are a few questions good enough? : UK Amicus Union Short Questionnaire: Are a few questions good enough? Yes No Q1) I work in a safe and healthy workplace Q2) I feel in control of my working environment Q3) My job is secure Q4) My job is interesting Q5) I am treated with fairness and dignity at work Q6) There is a trade union voice in my workplace Alternative ways of posing issue : Alternative ways of posing issue 1. Problems at work: factual rather than satisfied with – does employer make arbitrary decisions? Is there harassment? 2. Vignette approach – give different facts about job and ask people to rate them (Pouliakas, chapter 5) 3. Break measure into two parts: How much “influence” do you want/what is your expectation? How much do you have? (This interpretats of union impact: union improves pay/ benefits/ conditions but workers report lower job satisfaction and quit less. Why? They want more) Bryson-Freeman Needs/Problems: O/1 count of several objective variables : Bryson-Freeman Needs/Problems: O/1 count of several objective variables UK BWRPS needs (mean = 3.58; variance = 15.72) US WERS needs (mean = 3.08; variance = 11.39) The problem of Adaption : The problem of Adaption Kahneman: Cut off arm … happiness down … one yr later, happiness back to normal. Problem: Would you pay to get arm back? Pouliakas PhD: Estimated loss from going permanent to temporary job if currently perm -15.5% if currently temp -5.1% Inequality experience (ISSP): All countries –should be paid inequality is about 20% less than actual inequality regardless of level of inequality; over time in East Europe should rises as actual rises 3. Causes: People or workplaces? : 3. Causes: People or workplaces? Some individual factors – NLSY following people over time, get fixed effect. Follow 12,516 people over 18 years R-squared of 0.27 on individual fixed effects dummy variables Some of this is probably genetic related to whatever affects happiness; But socially interesting aspect is that job satisfaction is WORKPLACE related Evidence of Workplace Effects:Citicorps Bank Branches : Evidence of Workplace Effects:Citicorps Bank Branches 192 branches, many questions, 3482 workers, 18 items in this analysis, 2 years, 1994 and 1996 Branch effects have huge F stat 61.2 in ANOVA  “Happy and unhappy branches” Newly hired workers adopt the attitudes that the branches exhibited before they arrived Correlation of 1996 Branch Attitude on 1994 Attitude, Low Tenure vs Higher Tenure Workers : Correlation of 1996 Branch Attitude on 1994 Attitude, Low Tenure vs Higher Tenure Workers Evidence of Workplace Effects: WERS 1998 : Evidence of Workplace Effects: WERS 1998 NEEDS R-SQUARE FROM workplace dummies .203 R-SQUARE FROM demog + job measures .107 R-SQUARE FROM workplace dummies + demog + job measures .256 (Based on 25,451 workers in 1,759 workplaces). Needs measured from diverse questions; Bryson & Freeman Evidence of Workplace Effects WERS 2004 : Evidence of Workplace Effects WERS 2004 Satisfaction/Needs Scalar Measure based on 22 questions R-SQUARE from workplace dummies .203 Based on 22,451 workers in 1,732 workplaces 22 questions: 8 are satisfaction; 4 are about management informing workers; 10 trust management/management labor relations Person or Job Effect: ANOVA Results, NLSY79 1979-2000 : Person or Job Effect: ANOVA Results, NLSY79 1979-2000 Slide 28: 4. Does Job Changing raise satisfaction? What your partner/kids care about : Stressful jobs and Marital problems : What your partner/kids care about : Stressful jobs and Marital problems No studies of what happens to family situation when unhappy worker changes job Numerous studies: looks as if : Numerous studies: looks as if Marital Satisfaction  Job Satisfaction (Rogers, May 2003, others); Bad night of sleep  job dissatisfaction (same workers) Spouses view of job  job dissasfaction Job dissatisfaction  breakup/stress (5 + studies) Job-Work Conflict  Job satisfaction; household satisfaction (Kossek-Ozeki, 1998, meta analysis) 5. Consequences of Job Satisfaction : 5. Consequences of Job Satisfaction Lower turnover is universal finding: Most studies are “intent to leave” bcs of cross section data But we now have many longitudinal studies that show effect over time. No serious comparison of correls: if satisfaction-intent to leave correlation is ~ same as satisfaction actual leave correlation, would be strong support for stated choice method Less absences is also fairly universal Do Satisfied Workers/Workplaces Produce More? : Do Satisfied Workers/Workplaces Produce More? Moderate evidence for individuals but need meta-statistics to find it. Most recent analyses give larger impacts. Moderate evidence for workplaces Hypothesis: Stronger across workplaces than within Satisfied workplace  output Satisfied worker within unhappy workplace does not have much effect Happy Nonproductive Workers : Happy Nonproductive Workers Studies of Positive Affect and Creativity : Studies of Positive Affect and Creativity Ask people to give their state of mind at different times of day; someone rates their creativity over different time periods. Mixed bag of outcomes so people are looking for moderator variables Do you want to hire someone slightly weird and unhappy bcs they will be more creative? If they get happy will they lose creative drive? Some evidence that want people with schizophrenic relatives What Policies can  happier and more productive work place? : What Policies can  happier and more productive work place? US data suggest that “shared capitalist” modes of pay – profit-sharing/gain-sharing/employee stock ownership affects worker well-being and job satisfaction Also general finding that these systems  higher productivity. But no effort to see if job satisfaction is key intermediate variable But UK WERS data less uncertain that mode of pay affects worker well-being To improve job sat/productivity firms need to use employee opinion surveys more creatively : To improve job sat/productivity firms need to use employee opinion surveys more creatively 40% to 50% of US firms use opinion surveys, but they are not well used; treated as opinion poll Should treat them as diagnostic tool Treat work units as observation Need before/after on new employees get measure of attitudes before they start work Seek cause of problems: supervisor or workplace culture? Find specific questions and responses most related to output EXPERIMENT with workplace interventions The Worklife Wizard Research Effort : The Worklife Wizard Research Effort Begun by Dutch, expanded to many countries including UK US version focuses on workplace issues using module approach Inexpensive bcs web based, but problems of representativeness Provide reasonable benchmark for firms Working with to get the kind of questions that will help workers get the right job match

Add a comment

Related presentations