Widespread imprecision in estimates of the economic costs of invasive alien species worldwide

Abstract

Several hundred studies have attempted to estimate the monetary cost arising from the management and/or impacts of invasive alien species. However, the diversity of methods used to estimate the monetary costs of invasive alien species, the types of costs that have been reported, and the spatial scales at which they have been assessed raise important questions as to the precision of these reported monetary costs. Benford's Law has been increasingly used as a diagnostic tool to assess the accuracy and reliability of estimates reported in financial accounts but has rarely been applied to audit data on environmental costs. Therefore, the distributions of first, second- and leading double-digits of the monetary costs arising from biological invasions, as reported in the InvaCost database, were compared with the null expectations under Benford's Law. There was strong evidence that the reported monetary costs of biological invasions departed considerably from Benford's Law and the departures were of a scale equal to that found in global macroeconomic data. The rounding upwards of costs appears to be widespread. Furthermore, numerical heaping, where values cluster around specific numbers was evident with only 901 unique cost values accounting for half of the 13,553 cost estimates within the InvaCost database. Irrespective of the currency, the value of 1,000,000 was the most common cost estimate. An investigation of anomalous data entries concluded that non-peer reviewed official government reports need to provide greater detail regarding how costs are estimated. Despite the undeniably high economic cost of biological invasions worldwide, individual records of costs were often found to be imprecise and possibly inflated and this emphasises the need for greater transparency and rigour when reporting the costs of biological invasions. Identifying whether the irregularities found for the costs of biological invasions are general for other types of environmental costs should be a research priority.