RDP 2015-05: The Social Costs of Currency Counterfeiting 1. Introduction

Currency counterfeiting is costly for society: law enforcement agencies allocate resources to deter, detect and prosecute counterfeiting operations; households and businesses who mistakenly accept counterfeits as payment suffer a direct loss; and the wider public can lose confidence in the functionality of the currency. This is one reason why policymakers in Australia and around the world allocate substantial resources to the deterrence, detection and prosecution of counterfeiting activity. Nevertheless, little work has been done to quantify the social costs of counterfeiting despite its policy relevance.

The first part of this paper examines the social costs of counterfeiting and presents some key facts comparing the social costs in Australia to other economies. We find that the level of counterfeiting in Australia is relatively low compared with other economies, and that businesses incur a greater fraud loss from accepting counterfeits than households. We also discuss the costs incurred in counterfeiting prevention and the costs associated with diminished confidence in the currency; these two costs are likely to form a substantial proportion of the social costs of counterfeiting, but are difficult to measure.

Although the facts we present are informative, they do not provide information about the effects of counterfeiting on the demand for different methods of payment. This motivates the second part of this paper. We use a structural vector autoregression (VAR) to estimate the effect of an increase in counterfeiting activity on the demand for banknotes and close payment substitutes like debit cards and credit cards. We identify counterfeiting shocks in our structural model by assuming that these shocks take time to affect the demand for different methods of payment.^[1] This is motivated on the grounds that it takes time for the public to learn about the prevalence of counterfeiting and adjust their behaviour.

The model provides evidence that increased counterfeiting activity affects the payment methods chosen by the public. Based on the historical data, a one standard deviation increase in counterfeiting activity leads to a decline in the demand for banknotes of 0.20 per cent, which is consistent with some loss of confidence in currency. In response to the same counterfeiting shock, bank deposits increase by 0.04 per cent and credit card use increases by 0.16 per cent. These responses are consistent with substitution effects between methods of payment – specifically, a move away from currency and towards electronic means of payment.

We then estimate the increase in social costs due to differences in the costs of making payments electronically, as compared to using cash. Using transaction cost estimates from three Australian studies, we find that the average increase in social costs associated with the counterfeiting shock (where counterfeits detected increase by a cumulative total of around A$140,000 spread over a period of ten years) is A$7.0 million. The scale of this effect can be explained by confidence effects and the scale of the payments system compared to counterfeiting. Also, it should be noted that these estimates are subject to a degree of uncertainty.^[2]

This paper contributes to two areas of the literature. Our first contribution is to the small but growing literature about counterfeiting (see, for example, Chant (2004a), Fung and Shao (2011b) and Kim and Turton (2014)). Our second is to quantify the effects of counterfeiting on currency demand and alternative payment methods. This provides more information about the overall social costs of counterfeiting and is a new contribution. The main focus of existing empirical work has been to estimate the stock of counterfeits circulating from the level of counterfeits detected. Judson and Porter (2003) were the first to develop estimates for the United States, while Chant (2004b) and Bose and Das (2013) extend their methodologies to Canada and India. Another empirical paper examines some correlates of counterfeiting (Morris, Copes and Perry-Mullis 2009).

The rest of this paper is structured as follows. Section 2 discusses the channels through which counterfeiting affects the demand for currency. Section 3 presents some facts on the social costs of counterfeiting and compares the experience in Australia to other economies. Section 4 presents a structural model of counterfeiting and its effects on methods of payment. We also use this model to help quantify the overall social costs of counterfeiting. A range of robustness checks is explored in Section 5. Section 6 concludes.

Footnotes

Counterfeiting shocks as defined here are conceptually similar to counterfeiting attacks referred to in the literature. [1]

See Section 4.4 below for a detailed discussion. [2]