Stock crash and recovery: The Norwegian spring spawning herring

The collapse of the Norwegian spring spawning herring in the late 1960s is one of the most spectacular and well known fish stock collapses in the world. It is widely recognized that the collapse was due to a severe stock depletion in the wake of a sudden leap in technology, the consequences of which were not fully understood, together with open access exploitation by fleets from several nations. It took about twenty years for the stock to recover. We estimate a recruitment function for the herring stock from data beginning in the early 20th century, including the period of the collapse, and build a simulation model for the stock based on serially correlated random fluctuations around this recruitment function. In a deterministic model, fishing mortality rates in excess of 0.22 are not sustainable, but driving the stock to virtual extinction could be a matter of decades, for moderately excessive mortality rates. A rate of discount in excess of 23 percent could make such non-sustainable exploitation attractive, from an economic point of view, if the cost per unit of fish does not depend on the size of the stock. Random effects in recruitment can produce widely variable outcomes for the development of the stock, affecting the length of time it takes to deplete the stock to some reference level or its recovery from depletion. We investigate the effects of uncertainty in stock assessment, finding that it matters little, despite causing large variations in fishing mortality.