When are Real Options Exercised? An Empirical Study of Mine Closings

In this paper, we study a well-known real option: the opening and closing of mines. Using a new database that tracks the annual opening and closing decisions of 285 developed North American gold mines in the period 1988-1997, we confirm many of the predictions from real options models. In particular, the probability that a mine is open is related to market-wide factors (including the level and volatility of the gold price and the level of interest rates) as well as to mine-specific factors (including the mine's fixed costs, variable costs, and reserves.) These results are both statistically significant and economically material. There is strong evidence of hysteresis, which would result from non-zero opening and closing costs. Together, the data provide strong support for the real options model as a useful descriptor of mines' opening and shutting decisions. In addition, we find that the decision whether to shut a mine is related to firm-specific managerial factors not normally considered within a strict real options model, most notably the profitability of other mines in the firm's portfolio and of the firm's other businesses. These relationships do not seem to be based on geographic synergies and we suspect they are an indication of capital allocation processes within these firms