Financing Constraints and Fixed-Term Employment Contracts Andrea Caggese - Vicente Cuñat Universitat Pompeu Fabra
Intro Two important research topics Financing constraints (Macroeconomics, Corporate Finance) Fixed-term employment contracts (Labour Economics) Still quite incomplete knowledge on them Likely to interact Empirical analysis is especially important This paper aims to cover this gap
Financing constraints and investment Long literature on financing constraints and investment, both coming from Corporate Finance and Macroeconomics Some general theoretical results: excess volatility of working capital, low volatility and underinvestment on fixed capital More dynamic models yield more complex predictions than first generation (non-monotonicities) Empirical results under new scrutiny need revision
Financing constraints and employment Nickell and Nicolitsas (1999), Smolny and Winker (1999), Benito and Hernando (2003) [ ] Financing constraints increase the sensitivity of total employment to economic conditions Akin to literature on cash-flow/investment sensitivity and financing constraints Some empirical results show lower levels of total hiring of financially constrained firms
Employment: fixed-term, permanent contracts Bentolila and Saint Paul (1992), Bentolila and Bertola (1990), Hopenhayn and Rogerson (1993) among many others. Predictions on the relative use of fixed-term vs. permanent contracts and their volatilities Ambiguous predictions about total employment
Financing constraints fixed-term & permanent contracts Financing constraints do interact with employment through irreversibility restrictions and firing costs Fixed-term employment contracts relax these restrictions The co-existence of fixed-term employment contracts with permanent contracts is likely to generate interesting dynamics Largely an under-researched topic
This paper Shows the interaction between financing constraints and employment, taking into account the existence of fixed-term employment contracts Theoretical model: Dynamic employment decisions in the presence of financing constraints Fixed-term and permanent employment contracts coexist Permanent employment workers are more productive but subject to a firing cost Binding financing constraints matter, but expected financing constraints matter too
This paper (II) Calibrate the model to get additional predictions on employment use and volatilities Test empirically these additional predictions of the model
Summary of results Theory For calibrated parameter values, the model predicts that fixedterm workers are mainly used by financially fragile firms that want to to reduce future expected firing/hoarding labour costs. Financially constrained firms Use a higher proportion of fixed-term contracts Are more likely to use some fixed-term contracts Have more volatile fixed-term employment Have relatively less volatile permanent contracts Have more volatility of total employment
Model (I) We consider a risk neutral firm that maximises the discounted flow of dividends The firm chooses how much fixed-term and permanent workers to hire and the distribution policy (dividends vs. assets)
Model (II) Current wealth equals net income minus debt repayment Net income depends on the labour force and a productivity parameter Dividends + labour costs + firing costs = wealth + borrowing Where i tp takes a negative value equal to the amount of permanent workers fired
Model (III) Borrowing constraint Non negative dividends and maximum threshold on total borrowing
Model (IV) First order condition for fixed term workers: F.o.c. for permanent workers φ t is is the shadow value of financial wealth. It is zero for unconstrained firms. µ t Marginal cost of hoarding workers due to firing inactivity Ω t Expected costs of firing and hoarding permanent workers
Model (V) - When firms are hiring, the decision rule used to decide the type of contract of each marginal worker is: Where is a growing function of the amount of permanent workers hired. The ratio grows faster for constrained firms. - When firms are firing, they start firing fixed-term workers, then if more firing is needed they fire permanent workers until the firing cost equals the cost of hoarding excess workers.
Calibration Calibrate and Simulate an industry with many heterogeneous firms (100000 firm-year observations) Compare the statistics of constrained Vs. unconstrained firms Constrained firms: 50% of the simulated firms with the highest average value of λ (the Lagrange multiplier of the collateral constraint)
Calibration: Matching of empirical facts
Calibration: Predictions
Empirical Predictions More FT employment contracts by constrained firms Higher % of zero FT contracts by unconstrained ones Higher relative volatility of FT employment among constrained firms Higher volatility of FT employment for constrained Higher total volatility of employment for constrained firms
Data Mediocredito Italiano. Panel of small and medium sized Italian firms 6 valid years 1996-2000 16000 firm-year observations Restrict to firms with 10 or more employees Balance sheet, fixed-term and permanent employment Qualitative data on Firm not asking for credit for fear of denial Firm willing to accept higher interest rates for more credit Credit denial
Variables Financing constraints Size (above or below median) Self declared constraints (in at least one questions) 16% IV of Self declared constraints (IV s interest coverage, net liquid assets and regional financial development) Ratio of FT workers over permanent ones Coefficients of variation of ratio, total employment and individual magnitudes Control for volatility of sales, leverage, capital per worker, sector and year dummies
Constrained variable: Relationship with standard financing constraints variables constrained constrained First Stage Interest Coverage -0.085*** -0.010*** [0.005] [0.004] Net Liquid Assets/ Total Assets -0.123*** -0.083*** [0.019] [0.016] Financial Development Index -0.003*** -0.003*** [0.0002] [0.0002] log (Nummer of banks) 0.031*** [0.006] main bank located in same region -0.035*** [0.006] log(age) -0.006 [0.004] log(years with main bank) -0.011*** [0.003] Share of credit of main bank 0.0005*** [0.0001] Leverage -0.091*** -0.066*** [0.021] [0.018] log of Total Assets -0.028*** [0.003] Dummy if Sales Grow -0.001 [0.005] Change WIP Over Totalassets 0.113** [0.052] Fixed Capital Divided by Total Employment -0.0002*** [0.00003] sd (sales) -0.0000005*** [0.00000007] Excluded IV s R-squared of IV s 4% Total R-squared 7%
Large vs. Small Companies
Constrained variable
Constrained variable IV
Split Effect Size Self-Declared Self-Declared IV
Probability of FT>0
Conclusions Comprehensive approach to the interaction between financing constraints and FT employment contracts. Theoretical ambiguities clear up once the model is calibrated to fit actual data. Financing constraints lead to: Higher use and intensity of fixed-term contracts More volatile fixed-term employment Relatively less volatile permanent contracts More volatile total employment
Conclusions (II) Some policy implications: Employment and access to finance policies have to be considered jointly Employment security on permanent contracts affects volatility of fixed-term ones. This is emphasized by financing constraints New empirical evidence on the relevance of financing constraints