Abstract
Existing firm-dynamics models that quantify the role of entry in shaping aggregate fluctuations could not account for the following two salient features of the dynamics of entrants: First, entry rate is significantly procyclical and two times as volatile as aggregate employment. Second, cohort of firms that start operating during recessions employ fewer workers at entry and over time, while they are on average more productive and have higher survival rates compared to expansionary cohorts. Given the importance of start-ups for aggregate job creation and economic growth, the dissertation investigates what accounts for the observed significant and persistent effect of the initial entry conditions on the selection of entrants over the cycles and how does the observed life cycle demographics of entrants shape the aggregate fluctuations.
In the first part of the dissertation, I show that potential entrants’ ability to delay entry, previously ignored by the existing firm-dynamics literature, accounts for the significant effect of the aggregate conditions on selection of entrants. Procyclical variation in the expected survival rates moderate the relationship: during recessions increased risk of post-entry failure creates positive value of waiting and increases relative cost of entry today on top of fixed entry cost. The mechanism generates group of potential entrants who decide to postpone entry even if entry promises positive life-time value. I show that for reasonable parameter values (i) the total cost of entry during recessions increases by 7% for medium productivity entrants, (ii) the seemingly small increase in the entry cost generates delays that could last from 1 to 8 years. At an aggregate level the mechanism produces a countercyclical cost of entry which leads to the documented significant selection of entrants over the cycles. I show that without the ability to delay entry, existing firm dynamics models that rely on traditional entry decision rule and fixed entry cost require counterfactually large variance of the aggregate shock process to reconcile the observed facts.
In the second part of the dissertation, I investigate what accounts for the observed characteristics of cohorts over the cycles. I calibrate the model developed in previous section to U.S. establishment level data. The calibrated model matches average cohorts' post-entry survival, growth and employment share. The model is able to account for the observed dynamics of the entry rate and the documented persistent and significant differences in cohorts' life-cycle characteristics. I find that the option value of delay channel is quantitatively and qualitatively important for generating the results. Due to the option value of delay the opportunity cost of entry almost doubles during recessions, generating cohorts of entrants with countercyclical average productivity. The medium productivity entrants who delay entry contribute to the persistent procyclical variation in cohort-level employment, since they represent high-growth and high-survival firms. I find that the ability to delay entry allows potential entrants to choose time of entry that leads to higher profits for longer periods of time, leading to countercyclical survival rate. A model without the option value of delay implies acyclical average survival rate.
In the third part of the dissertation, I quantify the role the observed demographics of entrants play in shaping aggregate fluctuations. I find that a calibrated model that accounts for the life-cycle dynamics of U.S. establishments produces business-cycle fluctuations in aggregate employment and output that are similar to those observed in the data. Moreover, I show that the variation in the number and the composition of entrants at entry, rather than the post-entry shocks, is responsible for accounting for the observed persistence and variance of aggregate variables. Using the model, I re-examine the causal relationship between the persistent and significant drop in the number of entrants and the slow recovery observed after the Great Recession. A counterfactual exercise shows that if the entry rate had stayed at the pre-crisis level, the drop in aggregate employment would have been 45 percent lower, and the economy would have recovered two times faster.
In the final part of the dissertation, I show that accounting for the option to delay entry has important policy implications and could significantly alter the existing firm-dynamics models’ predictions about the response of potential entrants to various policies. In a model with the option value, a final effect of a policy depends on its effect on entrants' relative cost of entry today versus tomorrow, while the traditional entry decision rule accounts only for the direct effect of the policy. Depending on the magnitude of an indirect effect, these two specifications could imply quantitatively and qualitatively different responses of potential entrants depending on the timing, duration, and the magnitude of policies. For example, I show that existing models predict that a temporary or a permanent decrease in entry cost have a same effect on the number of entrants, whereas a model with the option value predicts that temporary decrease in entry cost is more effective in increasing the number of entrants during recessions since it decreases relative cost of entry more than the permanent policy.
