Abstract
We examine the correlates of reported hiring difficulties at the firm level using linked employer-employee and panel survey data over 2005-2011, focussing on the relative influence of firm-level characteristics, persistence, the business cycle and local labour market liquidity. At both the aggregate and the firm level, hiring difficulties eased after the onset of the Global Financial Crisis. Even in the presence of large cyclical changes in demand and labour market conditions, firm-level persistence is a dominant feature of the data, with one- and two-year lags of reported hiring difficulties both positively related to current difficulties. Firms paying higher wages are more likely to report difficulties when trying to hire skilled workers, while firms with more long tenure workers are less likely to report any difficulty hiring. Local labour market conditions appear unrelated to reported hiring difficulties.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The source and population that this statistic is derived from is explained in “Data”.
Cyclical changes in the search behaviour of firms over the cycle might also lead to observed differences in hiring outcomes.
Haskel and Martin (2001) extend this latter analysis to look at the relationship between reported skills shortages and technical change. Using the same data, Green et al. (1998) also explore the correlates of reported skills shortages, hard-to-fill vacancies and hiring difficulties, though their emphasis is on describing differences between these concepts.
Petrongolo and Pissarides (2001) review the matching function literature.
Fabling (2009) provides more detailed information on the content of the LBD.
Given the breadth of the survey content, Statistics New Zealand asks that the questionnaire be completed by the General Manager of the business, with “ …answers… confirmed with support people in specific areas.”
Most questions in BOS have low non-response/“don’t know” response rates, around the one-two percent level.
We lose a further 1.2 % of the sample because of missing employment information, and another 1.2 % to maintain a constant population definition over time.
All firm counts are random-rounded (base three) in concordance with Statistics New Zealand confidentiality rules.
Tests support the hypothesis that a “no difficulty” response corresponds to a firm posting vacancies that were easy to fill, over the alternative that such a response corresponds to a firm not posting a vacancy Fabling and Maré (2013).
LMAs are defined following Papps and Newell (2002), using the 58 LMA grouping, and reflect Census-based commute-to-work patterns.
We renormalise Maré et al.’s estimate of the firm wage premium to be zero-mean for the weighted population. These two-way fixed effects estimates were compiled on an earlier vintage of LEED, meaning that there are a very small number of firms for which the measure does not exist. For those firms, we set the premium to zero, and include an (unreported) dummy in regressions set to one for firms with missing data.
Pooling workers across years increases the number of firm-year observations for which this measure can be calculated. Despite this, there remain a small number of firms for which this measure cannot be derived either because the firm has not been an employer for four years in the LEED data, or because the firm has never had a worker with a two year tenure. The measure is set to zero for these firms and there is a separate (unreported) dummy variable, set to one for these observations, included in regressions.
LEED does not include hours worked, but the two-way fixed effect estimate makes an hours adjustment to the wages of workers who appear to be part-time.
Space constraints prevent a full enumeration of question wordings. However, BOS forms are available from Statistics New Zealand’s Questionnaires & Forms Database (http://www2.stats.govt.nz/domino/external/quest/sddquest.nsf).
Worker movements across LMAs, but within firms (eg, staff relocations), are excluded from flow measures because these may reflect reallocation of workers by Statistics New Zealand, rather than actual employee moves. In unreported results, we re-estimated all regressions with LMA-level variables constructed excluding multi-location firms. Results were very similar.
Firms in multiple regions have labour market variables that are additionally weighted by firm-level LMA employment shares.
Unlike the aggregate statistics in “Macro Trends”, these numbers are unweighted. Since large firms are more likely to report difficulty hiring and have lower survey weights (ie, are in fuller coverage strata), hiring difficulty rates are higher in unweighted statistics.
The exceptions are: for TAP, where transitions from no difficulty to no vacancy have a higher incidence rate; and for OTH, where transitions from severe to moderate difficulty are more likely.
Each firm-occupation group response is treated as a separate observation (ie, each firm contributes four observations in a year).
We examined whether firms changing state after the onset of the GFC were different from non-changers. Overall, there was not a strong systematic relationship between better performance and a weaker (or stronger) likelihood of changing state.
Since these statistics are conditional on posting a vacancy, we weight across the four occupational groups by the number of firms posting vacancies to construct the ranking. Hence, for a particular occupational group it is not necessarily the case that hiring difficulty rates are rank-ordered.
The table reports unweighted statistics, so the mean of the firm wage premium and long tenure share differ from the economy-wide averages reported in earlier tables.
In unreported regressions, we also included third lags of these variables. These display similar patterns of significance as the second period lags, with smaller coefficients.
Note that 2007 is the reference year throughout the paper, since this is the first year included in analysis with lagged hiring difficulties.
Fabling and Maré (2013) includes both sets of regression results in full.
These practices are only available in 2006 (BOS questions C17 and C19) and cover both the availability of flexible work arrangements (eg, job sharing, working from home) and other employment practices (eg, employee feedback, childcare). Each practice is represented by a binary set to one if the practice is held. For the purpose of the robustness test, firms non-wage practices are assumed to hold constant over all time. A further dummy was included and set to one for the 16 percent of firms not in the 2006 sample.
Alternatively, the selection of largely slow-moving or time-invariant firm characteristics imposes this relationship.
This finding is akin to that of Maré et al. (2014), who show that the positive relationship between local workforce characteristics and average firm innovation outcomes disappears in the presence of direct controls for firm practices.
Recalling that the “all other occupations” (OTH) employment share is the reference group, the fact that all reported employment share coefficients are negative in column 4 implies that firms with high OTH share have higher reported hiring difficulty of OTH workers than other firms.
References
Davis SJ, Faberman RJ, Haltiwanger JC (2012) Labor market flows in the cross section and over time. J Monet Econ 59:1–18
Davis SJ, Faberman RJ, Haltiwanger JC (2013) The establishment-level behavior of vacancies and hiring. Q J Econ 128:581–622
Fabling R (2009) A rough guide to New Zealand’s longitudinal business database. Global COE Hi-Stat Discussion Papers No. 103. Institute of Economic Research, Hitotsubashi University
Fabling R (2011) Keeping it together: tracking firms in New Zealand’s longitudinal business database. Working Paper 11-01. Motu Economic and Public Policy Research
Fabling R, Maré DC (2012) Cyclical labour market adjustment in New Zealand: the response of firms to the global financial crisis and its implications for workers. Working Paper 12-04. Motu Economic and Public Policy Research
Fabling R, Maré DC (2013) Firm-level hiring difficulties: persistence, business cycle and local labour market influences. Working Paper 13-06. Motu Economic and Public Policy Research
Green F, Machin S, Wilkinson D (1998) The meaning and determinants of skills shortages. Oxf Bull Econ Stat 60:165–87
Haskel J, Martin C (1993a) Do skill shortages reduce productivity? Theory and evidence from the United Kingdom. Econ J 103:386–394
Haskel J, Martin C (1993b) The causes of skill shortages in Britain. Oxf Econ Pap 45:573–88
Haskel J, Martin C (2001) Technology, wages, and skill shortages: evidence from UK micro data. Oxf Econ Pap 53:642–58
Hyslop D, Maré DC (2009) Job mobility and wage dynamics. LEED Research Paper. Statistics New Zealand
Lazear EP, Spletzer JR (2012) Hiring, churn, and the business cycle. Am Econ Rev 102:575–79
Maré DC, Fabling R, Stillman S (2014) Innovation and the local workforce. Pap Reg Sci 93:183–201
Maré DC, Hyslop D, Fabling R (2015) Firm productivity growth and skill. Working Paper 15–18. Motu Economic and Public Policy Research
Mok P, Mason G, Stevens P, Timmins J (2012) A good worker is hard to find: skills shortages in New Zealand Firms. Occasional Papers 12/5. Ministry of Economic Development
Moretti E (2011) Local labor markets. Elsevier. volume 4 of handbook of labor economics. chapter 14. 1237–1313
OECD (2005) OECD Employment Outlook. Organization for Economic Cooperation and Development, Paris
OECD (2014) Recruiting Immigrant Workers: New Zealand 2014. Organization for Economic Cooperation and Development, Paris
Papps KL, Newell JO (2002) Identifying functional labour market areas in New Zealand: a reconnaissance study using travel-to-work data. Discussion Paper 443. Institute for the Study of Labor (IZA)
Petrongolo B, Pissarides CA (2001) Looking into the black box: a survey of the matching function. J Econ Lit 39:390–431
Schank T, Schnabel C, Wagner J (2007) Do exporters really pay higher wages? First evidence from German linked employer-employee data. J Int Econ 72:52–74
Statistics New Zealand (2015) Labour Market Statistics: March 2015 quarter. Statistics New Zealand, Wellington
Venn D (2009) Legislation, Collective Bargaining and Enforcement: Updating the OECD Employment Protection Indicators. OECD Social, Employment and Migration Working Papers 89. OECD
Acknowledgments
Access to the data used in this study was provided by Statistics New Zealand in accordance with security and confidentiality provisions of the Statistics Act 1975 and the Tax Administration Act 1994. The results in this paper have been confidentialised to protect individual businesses from identification. See Fabling and Maré (2013) for the full disclaimer. This research was prepared with funding from the New Zealand Ministry of Business, Innovation and Employment. The authors thank the Ministry for its support, Statistics New Zealand for access to the data, and Sylvia Dixon, Lynda Sanderson, Jason Timmins, and participants at the Kyoto conference on “Employment adjustment in response to shocks” and CAED 2013 in Atlanta for feedback on the paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fabling, R., Maré, D.C. Firm-Level Hiring Difficulties: Persistence, Business Cycle And Local Labour Market Influences. J Labor Res 37, 179–210 (2016). https://doi.org/10.1007/s12122-016-9221-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12122-016-9221-8