\t <\/p>\n
20 September 2018<\/p>\n
\n\t\t\t\t\t\t\t\tlabour market, wages
\n\t\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t\t <\/p>\n
\t\t\t\t\t\t\t\t\t![\"\"\/](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/09\/bulletin-2018-09--09.jpg\")
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
<\/p>\n
\n\t\t\t\t\t\t\t\t\t\tPhoto: Sarote Pruksachat
\n\t\t\t\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\t\t
<\/p>\n
\t\t\t\t\t\t\tAbstract<\/p>\n
\n\t\t\t\t\t\t\t\tAustralia has a detailed system of \u2018awards\u2019 that specify different minimum
\n\t\t\t\t\t\t\t\twages depending on the industry, location and skill of an employee. I find
\n\t\t\t\t\t\t\t\tthat legislated adjustments to award wages in Australia between 1998 and
\n\t\t\t\t\t\t\t\t2008 were almost fully passed on to wages in award-reliant jobs. There is
\n\t\t\t\t\t\t\t\tno evidence that modest, incremental increases in award wages had an adverse
\n\t\t\t\t\t\t\t\teffect on hours worked or the job destruction rate.
\n\t\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe effects of minimum wages on employment have been widely debated internationally.
\n\t\t\t\t\t\t\tEconomic theory makes no clear predictions: although the traditional competitive
\n\t\t\t\t\t\t\tmodel of the labour market suggests that a minimum wage rise will reduce employment
\n\t\t\t\t\t\t\t\u2013 if set above the market clearing wage \u2013 other theoretical models
\n\t\t\t\t\t\t\tcan predict the opposite. For example, an increase in the minimum wage could
\n\t\t\t\t\t\t\tincrease employment in a labour market where firms have some degree
\n\t\t\t\t\t\t\tof market power that allows them to set wages (e.g. Card and Krueger (1994)).
\n\t\t\t\t\t\t\tEmpirical evidence for other countries has been varied, depending on the methodologies
\n\t\t\t\t\t\t\tand datasets used but, on balance, suggests that modest and incremental increases
\n\t\t\t\t\t\t\tin minimum wages do not have significant adverse effects on hours worked and
\n\t\t\t\t\t\t\tjob loss. There has been limited empirical evidence for Australia, partly
\n\t\t\t\t\t\t\tbecause of data constraints. This article contributes to the debate by using
\n\t\t\t\t\t\t\tan approach and a job-level dataset that are uniquely suited to studying the
\n\t\t\t\t\t\t\teffects of minimum wage increases in Australia.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe analysis focuses on the minimum wage changes most common in Australia: annual
\n\t\t\t\t\t\t\tadjustments by the Fair Work Commission
\n\t\t\t\t\t\t\t(FWC).
\n\t\t\t\t\t\t\tThese adjustments have typically been modest, incremental and, to some extent, predictable.
\n\t\t\t\t\t\t\tThe analysis finds that these changes are almost fully passed on to wages
\n\t\t\t\t\t\t\tin award-reliant jobs, and appear to have had little adverse effect on hours
\n\t\t\t\t\t\t\tworked or job loss. These findings are consistent with the international evidence
\n\t\t\t\t\t\t\tand the FWC’s (2018) current assessment of that evidence. However, the
\n\t\t\t\t\t\t\tfindings do not necessarily generalise to large, unanticipated changes in
\n\t\t\t\t\t\t\tminimum wages.
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t
\n\t\t\t\t\t\t\tMinimum Wages in Australia
<\/p>\n
\n\t\t\t\t\t\t\tThe National Minimum Wage (NMW) sets a legal floor on wages. As of 1 July 2018, this
\n\t\t\t\t\t\t\twas equal to $18.93 per hour for an adult employee. Australia also has a detailed
\n\t\t\t\t\t\t\tsystem of award wages that are layered on top of the NMW. These \u2018award
\n\t\t\t\t\t\t\twages\u2019 can depend on the industry, age, skill level and qualifications
\n\t\t\t\t\t\t\tof an employee, among other factors. For example, an employee who supervises
\n\t\t\t\t\t\t\tlift operators at a ski resort is entitled to a wage of $24.63 per hour, 30\u00a0per\u00a0cent
\n\t\t\t\t\t\t\tabove the NMW. As such, while many \u2018minimum wage\u2019 employees are
\n\t\t\t\t\t\t\tpaid the NMW \u2013 particularly lower-skilled workers and those not covered
\n\t\t\t\t\t\t\tby an award or enterprise agreement \u2013 many others are entitled to a
\n\t\t\t\t\t\t\thigher wage.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe NMW and all award wages are adjusted at the same time every year, usually in
\n\t\t\t\t\t\t\tearly July. The size of these adjustments is decided at a national level by
\n\t\t\t\t\t\t\tthe FWC. Any adjustments are applied consistently across awards; historically
\n\t\t\t\t\t\t\tthe FWC has either added a flat dollar amount to all award wages (1993\u20132010)
\n\t\t\t\t\t\t\tor raised all award wages by the same percentage amount (2011\u201318). Adjustments
\n\t\t\t\t\t\t\tto individual awards are less common. In its recent Annual Wage Review, the
\n\t\t\t\t\t\t\tFWC announced a 3.5\u00a0per\u00a0cent increase in the NMW and minimum wages
\n\t\t\t\t\t\t\tacross all awards, which took effect on 1 July 2018 (Graph 1).
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
<\/p>\n
Graph 1<\/p>\n
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t![\"Graph](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/09\/graph-0918-9-01.gif\")
<\/p>\n
\n\t\t\t\t\t\t\tAustralia’s system for setting minimum wages is more complicated than in most
\n\t\t\t\t\t\t\tother countries. Unlike other countries that set a single minimum wage (e.g.
\n\t\t\t\t\t\t\tthe United Kingdom, Germany and New Zealand) or a minimum wage that varies
\n\t\t\t\t\t\t\tby state (e.g. the United States), the Australian system has thousands of
\n\t\t\t\t\t\t\taward wages that currently range from $18.93 per hour to as high as $171.00
\n\t\t\t\t\t\t\tper hour (Graph 2). Because award wages are often set above the NMW, the share
\n\t\t\t\t\t\t\tof Australian employees whose pay is set according to a minimum wage is relatively
\n\t\t\t\t\t\t\thigh by world standards. Nearly one-quarter of all employees in Australia
\n\t\t\t\t\t\t\thave their wage set according to a minimum or award wage, and this share has
\n\t\t\t\t\t\t\trisen over recent years (RBA (Reserve Bank of Australia) 2017). The large
\n\t\t\t\t\t\t\tshare of employees affected by FWC decisions also reflects that the NMW in
\n\t\t\t\t\t\t\tAustralia is high relative to other nations, both in absolute terms and as
\n\t\t\t\t\t\t\ta share of median earnings (Productivity Commission 2015a).
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
<\/p>\n
Graph 2<\/p>\n
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t![\"Graph](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/09\/graph-0918-9-02.gif\")
<\/p>\n
\t\t\t\t\t\t
\n\t\t\t\t\t\t\tThe Effect of Minimum Wages on Employment
<\/p>\n
\n\t\t\t\t\t\t\tAustralia’s unique institutions for setting minimum wages mean that the findings
\n\t\t\t\t\t\t\tof empirical studies for other countries do not necessarily apply in the Australian
\n\t\t\t\t\t\t\tcontext. The complexity in Australia’s wage-setting institutions also
\n\t\t\t\t\t\t\tmakes estimating the effects of minimum wages using Australian data a difficult
\n\t\t\t\t\t\t\ttask.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe main challenge in estimating the effect of minimum wages on employment is that
\n\t\t\t\t\t\t\tit is hard to disentangle the effects of minimum wages from the effects of
\n\t\t\t\t\t\t\teverything else that is going on in the labour market. For example, if minimum
\n\t\t\t\t\t\t\twages are adjusted at the same time as a change in income tax rates, then
\n\t\t\t\t\t\t\tany change in employment that occurs around the change in minimum wages may
\n\t\t\t\t\t\t\tin fact be due to the tax changes rather than the FWC decision. To isolate
\n\t\t\t\t\t\t\tthe effects of the minimum wage changes on employment, we need to consider
\n\t\t\t\t\t\t\ta group of workers who are affected by the change and another group that are
\n\t\t\t\t\t\t\tnot (i.e. a control group), but are otherwise subject to the same forces affecting
\n\t\t\t\t\t\t\tthe economy and the labour market. The challenge for researchers is that it
\n\t\t\t\t\t\t\tcan be difficult to find a suitable control group of individuals who are not
\n\t\t\t\t\t\t\taffected \u2013 or less affected \u2013 by the change.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tI overcome this challenge by recognising that, while all jobs have their minimum
\n\t\t\t\t\t\t\twages adjusted at the same time each year, historically, some jobs have had
\n\t\t\t\t\t\t\tlarger adjustments than others. This means that we can compare the labour
\n\t\t\t\t\t\t\tmarket outcomes of those receiving a relatively large adjustment to those
\n\t\t\t\t\t\t\treceiving a smaller adjustment.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe decade leading up to the late 2000s is particularly useful for studying the effects
\n\t\t\t\t\t\t\tof award wages on labour market outcomes. During this period, the FWC routinely
\n\t\t\t\t\t\t\tgranted flat dollar increases to all awards each year, irrespective of the
\n\t\t\t\t\t\t\texisting wage rates contained in those awards. For example, in May 2000, the
\n\t\t\t\t\t\t\tFWC increased all award wages by 39 cents per hour. This meant that employees
\n\t\t\t\t\t\t\ton relatively low award wages received larger wage increases in percentage
\n\t\t\t\t\t\tterms than those on higher award wages: a worker earning $10.14 per hour
\n\t\t\t\t\t\t\t(the NMW at the time) was entitled to a wage rise of 3.9\u00a0per\u00a0cent,
\n\t\t\t\t\t\t\twhile a worker on $18 per hour received an increase of only around 2\u00a0per\u00a0cent
\n\t\t\t\t\t\t\t(Graph 3). On some occasions, the FWC also granted smaller flat dollar increases
\n\t\t\t\t\t\t\tto higher award wage ranges, rather than a single flat dollar increase. For
\n\t\t\t\t\t\t\texample, in 2003, the FWC announced an increase of 45 cents per hour for all
\n\t\t\t\t\t\t\taward wages up to $19.26 per hour, and 39 cents per hour for all wages above
\n\t\t\t\t\t\t\tthis level. This created a discontinuity in the profile of percentage wage
\n\t\t\t\t\t\t\tincreases at $19.26 per hour (Graph 3). These discontinuities tended to exacerbate
\n\t\t\t\t\t\t\tthe differences in the size of the award wage increase between low- and high-award-wage
\n\t\t\t\t\t\t\temployees.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThis article uses these differences in the size of minimum wage adjustments across
\n\t\t\t\t\t\t\tdifferent jobs to disentangle the separate effect of minimum wages from the
\n\t\t\t\t\t\t\teffect of other changes in the labour market. The idea is to simply compare
\n\t\t\t\t\t\t\tthe change in each variable of interest \u2013 wages, hours worked or job
\n\t\t\t\t\t\t\tloss \u2013 around each FWC decision between jobs that experienced a relatively
\n\t\t\t\t\t\t\tlarge percentage adjustment and those that only had a small adjustment.
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
<\/p>\n
Graph 3<\/p>\n
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t![\"Graph](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2025\/09\/graph-0918-9-03.gif\")
<\/p>\n
\t\t\t\t\t\t
\n\t\t\t\t\t\t\tData
<\/p>\n
\n\t\t\t\t\t\t\tThe approach described above requires detailed data on the wages and employment outcomes
\n\t\t\t\t\t\t\tof individual employees or individual jobs before and after each FWC decision.
\n\t\t\t\t\t\t\tThis article uses a source of data not previously used in Australian research
\n\t\t\t\t\t\t\ton minimum wages: job-level data from the Australian Bureau of Statistics
\n\t\t\t\t\t\t\t(ABS) Wage Price Index (WPI) survey. This survey includes 3,000 firms per
\n\t\t\t\t\t\t\tquarter. Each firm is being surveyed every quarter for five years, before
\n\t\t\t\t\t\t\tbeing rotated out of the sample (roughly one-fifth of the sample is replaced
\n\t\t\t\t\t\t\teach year). After being selected into the sample, a firm is asked to randomly
\n\t\t\t\t\t\t\tselect a certain number of jobs from their payroll records. The firm then
\n\t\t\t\t\t\t\treports information on each of these jobs over time. Around 18,000 jobs are
\n\t\t\t\t\t\t\tincluded in the survey every quarter. Approximately 15\u201320\u00a0per\u00a0cent
\n\t\t\t\t\t\t\tof these jobs have their pay set exactly according to an award.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tA key advantage of this dataset is that it includes both a precise measure of actual
\n\t\t\t\t\t\t\twages and an indicator for whether a job’s wage is set according to a
\n\t\t\t\t\t\t\tminimum or award wage. This allows one to focus on jobs directly affected
\n\t\t\t\t\t\t\tby minimum wage decisions. This approach differs from most previous studies
\n\t\t\t\t\t\t\t(particularly for the United States), which, due to data limitations, tend
\n\t\t\t\t\t\t\tto focus on low-wage industries, such as the restaurant sector, or on lower-productivity
\n\t\t\t\t\t\t\temployees, such as teenagers. Using low-wage groups as a proxy for employees
\n\t\t\t\t\t\t\treliant on minimum wages may lead to biased estimates of the effect of minimum
\n\t\t\t\t\t\t\twages on employment (Jardim et al 2017). The WPI data also allow one to study the effect
\n\t\t\t\t\t\t\tof changes to minimum wages on hourly wages themselves, which is rarely possible
\n\t\t\t\t\t\t\twith more traditional data. The pass-through of award wages to hourly wages
\n\t\t\t\t\t\t\tis a key consideration for the FWC in terms of assessing whether firms are
\n\t\t\t\t\t\t\tcomplying with their legal obligations. The elasticity of hourly wages with
\n\t\t\t\t\t\t\trespect to minimum wages is also one of the key parameters in assessing the
\n\t\t\t\t\t\t\teffect of minimum wage increases on income inequality (Leigh 2007).
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe WPI survey follows jobs, rather than employees (for example, a graduate economist
\n\t\t\t\t\t\t\tat the Reserve Bank of Australia). If the occupant of the job leaves the firm
\n\t\t\t\t\t\t\tor moves to a different job within the firm, the ABS substitutes the job leaver
\n\t\t\t\t\t\t\twith the employee who replaced her or an existing employee with the same job
\n\t\t\t\t\t\t\ttitle. For this reason, the results in this study pertain to jobs, rather
\n\t\t\t\t\t\t\tthan individual employees. The estimation sample includes all private sector
\n\t\t\t\t\t\t\tjobs filled by an adult on an award rate. Juniors, apprentices and trainees
\n\t\t\t\t\t\t\tare excluded from the analysis, as their award wage adjustment cannot be inferred. This is unfortunate
\n\t\t\t\t\t\t\tbecause these groups may be particularly vulnerable to job loss following
\n\t\t\t\t\t\t\tan increase in award
\n\t\t\t\t\t\t\twages. This leaves a sample
\n\t\t\t\t\t\t\tof 32,174 job-period observations spanning the 11 FWC decisions over the period
\n\t\t\t\t\t\t\tbetween 1998 and 2008.
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t
\n\t\t\t\t\t\t\tEmpirical Strategy
<\/p>\n
\n\t\t\t\t\t\t\tA difference-in-differences (DD) model is used to estimate the effects of minimum
\n\t\t\t\t\t\t\twage changes on wages, hours worked and the job destruction rate of award-reliant
\n\t\t\t\t\t\t\tjobs. The intuition for this approach was discussed above; the DD model compares
\n\t\t\t\t\t\t\tthe change in each outcome variable \u2013 wages, hours worked or job destruction
\n\t\t\t\t\t\t\t\u2013 around each FWC decision between jobs that experienced a relatively
\n\t\t\t\t\t\t\tlarge percentage increase in their award wage and those that experienced a
\n\t\t\t\t\t\t\trelatively small increase. This means jobs that experienced a relatively small
\n\t\t\t\t\t\t\tadjustment to their award wages fulfil the role of a \u2018control group\u2019
\n\t\t\t\t\t\t\tfor those jobs that had a larger increase. The size of the award wage adjustment
\n\t\t\t\t\t\t\ta particular job received is inferred from its wage level immediately before
\n\t\t\t\t\t\t\tthe decision. Given that attention is restricted to jobs paid exactly an award
\n\t\t\t\t\t\t\twage before each FWC decision, the results provide direct estimates of the
\n\t\t\t\t\t\t\teffects of minimum wages on award-reliant jobs.
\n\t\t\t\t\t\t<\/p>\n
\t\t\t\t\t\t
\t\t\t\t\t\t
\t\t\t\t\t\t\t
<\/p>\n Sources: ABS; RBA<\/p>\n
\t\t\t\t\t\t
\t\t\t\t\t\t
\t\t\t\t\t\t
\t\t\t\t\t\t\t\t
\t\t\t\t\t\t\t
\t\t\t\t\t\t
\t\t\t\t\t\t
\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\tChanges in wages, hours worked or job destruction around each FWC decision are measured
\n\t\t\t\t\t\t\tfrom the WPI survey immediately before the decision to the survey six months
\n\t\t\t\t\t\t\tlater. This gives valid estimates of the effect of the minimum wage on wages,
\n\t\t\t\t\t\t\thours worked and job loss to the extent that employers’ adjustments to
\n\t\t\t\t\t\t\taward changes happen within six months (Borland 2018). During the period between
\n\t\t\t\t\t\t\t1998 and 2008, there were 11 minimum wage decisions. The approach taken here
\n\t\t\t\t\t\t\tis to pool the 11 decisions together and use a single DD estimator that constrains
\n\t\t\t\t\t\t\tthe coefficient of interest to be constant across the decisions. This maximises
\n\t\t\t\t\t\t\tthe available sample to estimate the effects of interest. The model includes
\n\t\t\t\t\t\t\tcontrols for any macroeconomic shocks that affect all wage groups and controls
\n\t\t\t\t\t\t\tfor any characteristics of each wage group that do not change over
\n\t\t\t\t\t\t\ttime.
\n\t\t\t\t\t\t\tDetails of the model are provided in Appendix A<\/a>.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThe following outcome variables are considered:
\n\t\t\t\t\t\t<\/p>\n
\n
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\tWages: the log of the job’s hourly wage, excluding any
\n\t\t\t\t\t\t\t\twage changes due to changes in the job occupant’s grade or performance.
\n\t\t\t\t\t\t\t
\n
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\tHours worked: the log of the ordinary-time hours paid for
\n\t\t\t\t\t\t\t\tduring the most recent pay period. This includes hours of paid leave (e.g.
\n\t\t\t\t\t\t\t\tannual leave and sick leave) but excludes overtime hours.
\n\t\t\t\t\t\t\t
\n
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\tJob destruction rate: a binary variable that equals one
\n\t\t\t\t\t\t\t\tif the job had ceased by the survey date in question, conditional on the
\n\t\t\t\t\t\t\t\tjob existing six months earlier, and zero otherwise. This captures both redundancies
\n\t\t\t\t\t\t\t\tand firm failure, but does not include cases where the job incumbent leaves
\n\t\t\t\t\t\t\t\tthe firm (either voluntarily or involuntarily) or if the firm only makes
\n\t\t\t\t\t\t\t\tsome of the employees in a given position redundant.
\n\t\t\t\t\t\t\t
\n
<\/p>\n
\n\t\t\t\t\t\t\tThe estimates for wages and hours worked are conditional on the job being in the
\n\t\t\t\t\t\t\tsample in the survey before and after a given FWC decision. If a job is vacant
\n\t\t\t\t\t\t\tor made redundant in either period, it is dropped from the sample for estimating
\n\t\t\t\t\t\t\tthe wage and hours effect. This is not the case for the job destruction rate
\n\t\t\t\t\t\t\testimate.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tResults
<\/p>\n
\n\t\t\t\t\t\t\tThe estimates for wages, hours worked and the job destruction rate are shown in Table
\n\t\t\t\t\t\t\t1. Separate estimates are presented using the full sample and a sample that
\n\t\t\t\t\t\t\tomits the first four FWC decisions (1998\u20132001). Although the sample
\n\t\t\t\t\t\t\tsize is larger for the full sample, there is a data issue prior to 2002 that
\n\t\t\t\t\t\t\tcould potentially distort the regression
\n\t\t\t\t\t\t\tresults.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\tTable 1: Effect of Award Wages on Wages, Hours Worked and the Job Destruction Rate(a)
<\/p>\n
\n\t\t\t\t\t\t\t\t\t\tEffect of a 1\u00a0per\u00a0cent increase in award wages
\n\t\t\t\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\u00a0
\n\t\t\t\t\t\t\t\t\t\tWages
\n\t\t\t\t\t\t\t\t\t\tHours worked
\n\t\t\t\t\t\t\t\t\t\tJob destruction rate
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\u00a0
\n\t\t\t\t\t\t\t\t\t\t(%)
\n\t\t\t\t\t\t\t\t\t\t(%)
\n\t\t\t\t\t\t\t\t\t\t(ppt)
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t1998\u20132008 sample
\n\t\t\t\t\t\t\t\t\t\t0.84***
\n\t\t\t\t\t\t\t\t\t\t0.26
\n\t\t\t\t\t\t\t\t\t\t\u22120.22
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\u00a0
\n\t\t\t\t\t\t\t\t\t\t(0.03)
\n\t\t\t\t\t\t\t\t\t\t(0.37)
\n\t\t\t\t\t\t\t\t\t\t(0.44)
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t2002\u201308 sample
\n\t\t\t\t\t\t\t\t\t\t0.93***
\n\t\t\t\t\t\t\t\t\t\t0.13
\n\t\t\t\t\t\t\t\t\t\t\u22120.37
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\u00a0
\n\t\t\t\t\t\t\t\t\t\t(0.04)
\n\t\t\t\t\t\t\t\t\t\t(0.47)
\n\t\t\t\t\t\t\t\t\t\t(0.55)
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t
<\/p>\n
\n\t\t\t\t\t\t\t\t\t\t\t\t(a) Standard errors (in parentheses) are clustered at the individual job level; ***,
\n\t\t\t\t\t\t\t\t\t\t\t\t** and * denote statistical significance at the 1, 5 and 10\u00a0per\u00a0cent
\n\t\t\t\t\t\t\t\t\t\t\t\tlevels, respectively
\n\t\t\t\t\t\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\t
<\/p>\n
\n\t\t\t\t\t\t\tThere is strong evidence that award adjustments are passed on to wages. Using the
\n\t\t\t\t\t\t\tfull sample, the elasticity of wages with respect to the award wage is estimated
\n\t\t\t\t\t\t\tto be 0.84 and highly statistically significant (Table 1). The estimate is
\n\t\t\t\t\t\t\tlarger in the shorter sample period (0.93) which suggests that measurement
\n\t\t\t\t\t\t\tissues in the first few years of data may be attenuating the size of the overall
\n\t\t\t\t\t\t\telasticity in the full sample. A wage elasticity of 0.93 implies that a 1\u00a0per\u00a0cent
\n\t\t\t\t\t\t\tincrease in award wages leads to a 0.93\u00a0per\u00a0cent increase in actual
\n\t\t\t\t\t\t\twages for award-reliant workers. Full pass-through of minimum wage changes
\n\t\t\t\t\t\t\tinto wages would require this elasticity be equal to one. Given the estimate
\n\t\t\t\t\t\t\tis slightly less than one, it suggests less-than-full pass-through of FWC
\n\t\t\t\t\t\t\tdecisions to wages. This may reflect
\n\t\t\t\t\t\t\tsome degree of noncompliance by firms in their legal obligations (see Productivity
\n\t\t\t\t\t\t\tCommission (2015b, p 12) for a discussion), or simply measurement error. In
\n\t\t\t\t\t\t\tany case, the elasticity is sufficiently close to one to characterise this
\n\t\t\t\t\t\t\tas near-complete pass-through.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tHaving established that award wage increases are passed through to wages, the next
\n\t\t\t\t\t\t\tstep is to see if this leads to any reduction in hours worked or job loss;
\n\t\t\t\t\t\t\tthis would be the case if the DD estimates are negative for hours worked and
\n\t\t\t\t\t\t\tpositive for the job destruction rate. There is no evidence that award wage
\n\t\t\t\t\t\t\tchanges have an adverse effect on hours worked or the job destruction rate:
\n\t\t\t\t\t\t\tin both cases the DD estimate is not significantly different from zero (Table
\n\t\t\t\t\t\t\t1). (And the point estimates were in fact the opposite signs to an adverse
\n\t\t\t\t\t\t\teffect.)
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\tRobustness tests
<\/p>\n
\n\t\t\t\t\t\t\tLike any analysis using a DD approach, these results are sensitive to violations
\n\t\t\t\t\t\t\tof the so called \u2018parallel trends\u2019 assumption. The assumption
\n\t\t\t\t\t\t\tis that, in the absence of a change in award wages, the wages, hours worked
\n\t\t\t\t\t\t\tand probability of job loss of low-award-wage workers would have followed
\n\t\t\t\t\t\t\tthe same trajectory as high-award-wage workers. This would be violated if,
\n\t\t\t\t\t\t\tsay, there is a change in tax policy that affects low-wage earners relatively
\n\t\t\t\t\t\t\tmore than high-wage earners that occurs at the same time as the change in
\n\t\t\t\t\t\t\taward wages. If this happens, we may incorrectly attribute the effect of the
\n\t\t\t\t\t\t\ttax change to the award wage. Bishop (2018) discusses several ways of testing
\n\t\t\t\t\t\t\tthe validity of the parallel trends assumption. These include placebo tests,
\n\t\t\t\t\t\t\tusing additional control groups, controlling for firm-specific shocks, and
\n\t\t\t\t\t\t\taccounting for \u2018pre trends\u2019 in the outcome variables. The results
\n\t\t\t\t\t\t\tof these tests suggest that the baseline estimates are likely to be robust
\n\t\t\t\t\t\t\tto potential violations of the parallel trends assumption.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tConclusion
<\/p>\n
\n\t\t\t\t\t\t\tThere is widespread interest in understanding the effects of minimum wage increases.
\n\t\t\t\t\t\t\tThis article adds to the evidence base by using an approach uniquely suited
\n\t\t\t\t\t\t\tto Australia and a dataset that provides several advantages over those used
\n\t\t\t\t\t\t\tpreviously in the literature. It finds that small, incremental adjustments
\n\t\t\t\t\t\t\tto awards are mostly passed on to wages in award-reliant jobs. These adjustments
\n\t\t\t\t\t\t\tappear to have no discernible adverse effect on hours worked or job loss.
\n\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\tThere are several things to keep in mind when interpreting these findings. Firstly,
\n\t\t\t\t\t\t\tas discussed, these results are for adult employees only and do not include
\n\t\t\t\t\t\t\tjuniors. Secondly, the results do not necessarily generalise to large, unanticipated
\n\t\t\t\t\t\t\tchanges in award wages. There will always be some point at which a minimum
\n\t\t\t\t\t\t\twage adjustment will begin to reduce employment significantly. Thirdly, the
\n\t\t\t\t\t\t\tapproach considers six month windows around FWC decisions; it is possible
\n\t\t\t\t\t\t\tthat firms take longer to respond to changes in minimum wages. Finally, although
\n\t\t\t\t\t\t\tthis article finds no evidence of an effect of award adjustments on job destruction,
\n\t\t\t\t\t\t\tthis does not rule out an adverse effect on employment. It is possible that
\n\t\t\t\t\t\t\tthe adverse consequences of higher wage floors may be borne by job seekers,
\n\t\t\t\t\t\t\trather than current job holders.
\n\t\t\t\t\t\t<\/p>\n
\n
<\/p>\n
\n\t\t\t\t\t\t\t\t\tAppendix A The Difference-in-differences Model
<\/p>\n
\n\t\t\t\t\t\t\t\t\tThe DD equation is,
\n\t\t\t\t\t\t\t\t<\/p>\n
\n\t\t\t\t\t\t\t\t\twhere yikt is the dependent variable of interest for job i
\n\t\t\t\t\t\t\t\t\tin wage group k at time t. \u03bbt
\n\t\t\t\t\t\t\t\t\tis a full set of time dummies that control for any macroeconomic shocks
\n\t\t\t\t\t\t\t\t\tthat affect all wage groups in any of the 22 time periods (there are 11
\n\t\t\t\t\t\t\t\t\tdifferent FWC decisions each with their own \u2018before\u2019 and \u2018after\u2019
\n\t\t\t\t\t\t\t\t\tperiod).
\n\t\t\t\t\t\t\t\t\tFWCk is the log change in award wages for
\n\t\t\t\t\t\t\t\t\twage group k
\n\t\t\t\t\t\t\t\t\tdue to the FWC decision. There are also a set of interactions between
\n\t\t\t\t\t\t\t\t\tFWCk and dt, the latter being a
\n\t\t\t\t\t\t\t\t\tdummy that takes the value of zero in the \u2018before\u2019 period immediately
\n\t\t\t\t\t\t\t\t\tprior to an award increase and a value of one in the \u2018after\u2019
\n\t\t\t\t\t\t\t\t\tperiod six months later. The coefficient of interest is \u03b23.
\n\t\t\t\t\t\t\t\t\tWhen the dependent variable is the log hourly wage (or log hours worked),
\n\t\t\t\t\t\t\t\t\tthis DD coefficient is the elasticity of wages (or hours) with respect to
\n\t\t\t\t\t\t\t\t\tthe award wage, a parameter of key interest to policymakers.
\n\t\t\t\t\t\t\t\t<\/p>\n
\n
<\/p>\n
\n
\n\t\t\t\t\t\t\t
\n
<\/p>\n
\n
<\/p>\n
\n\t\t\t\t\t\t\t\t\tReferences
<\/p>\n
\n\t\t\t\t\t\t\t\t\tABS (Australian Bureau of Statistics) (2017), \u2018Employee Earnings and Hours,
\n\t\t\t\t\t\t\t\t\tAustralia, May 2016\u2019, ABS Cat No 6306.0, January.
\n\t\t\t\t\t\t\t\t<\/p>\n