Simultaneous and Solitary Spikes: Investment in Buildings and Equipment

We study firm-level expansionary investment activities in both equipment and buildings – so-called investment spikes. Our identification strategy decomposes firm investment spikes into three streams: a spike in equipment only, buildings only or a simultaneous spike. Empirically, we find that the timing and size of investment in equipment and buildings are not independent. Firms conducting a simultaneous spike enhance firm scale more than in the case of a spike in equipment or buildings alone. Employment growth occurs when a firm builds structures. Investment in equipment affects the optimal input mix and high productivity in equipment and buildings provides investment timing signals. In low-tech sectors firm production growth depends on investment in buildings. In contrast, a necessary condition for firms in high-tech sectors to grow their production is investment in equipment.

This paper explores the consequences of firm level investment spikes in productive capital, like equipment, and non-productive capital, like buildings. Our contribution is to examine how the composition of firms’ investment spikes affects the scale of production and employment, productivity, the input mix and operational efficiency. Buildings are an important production factor; they house employees and shield equipment. In our analysis, we investigate whether investment in structures drives employment, production technology and firm capacity in manufacturing industries, and also distinguish industries by research and labour intensity. We find that identifying investment spikes in buildings and equipment has implications for the productivity literature and principally, how we measure the extensive and intensive margins of productivity. Our empirical results document to us that investment in buildings and equipment is interrelated – the timing and size of investment in equipment and buildings are not independent phenomena. We also find that adding investments in buildings to a firm’s decision set improves understanding of key firm level performance and production metrics.

Depicts the production scale and number of workers before, during and after the investment spike. The vertical axis represents the percentage difference relative to firms that experienced no spike event. Markers represent estimates significant at p < .10 level (two-tailed).

Moreover, the decomposition further calibrates an understanding of the intensive margin. A measure of the investment size is more informative when including both expenditures on buildings and equipment. Our empirical results document that firms who signal expansion through simultaneous investment spikes in both buildings and equipment experience a higher post investment expansion in production and number of workers, than firms that experience a spike in either equipment or buildings only. However, the results also reveal that large investments do not improve firm level productivity. Instead high-productivity acts as a signal of when to invest, where before an investment takes place firm productivity is high and afterwards it decreases.

 

Finally, our empirical study highlights that production process mixes between building and equipment impact productivity and employees different and are fundamentally different across industry sectors. Our results suggest investment in equipment tends to increase the employee wage rate at a firm on average; based on this result we infer that firms buying new machinery display an increase in the skilled worker ratio. Likewise we deduce firms investing in structures hire more unskilled workers. Furthermore, when firms invest in equipment, the labour intensity decreases as well. These latter findings suggest that capital investments also affect the production technology employed by the firm. In addition, we show firm investments affect operational efficiency. Firms in high-tech sectors rely more on investment in equipment to be able to grow, whereas companies in low-tech sectors need investment in buildings to be able to expand.