چه زمانی صورت حساب های پلاستیکی صورتحساب بانک مرکزی را کاهش می دهند؟ مدل و برآوردهای برای ایالات متحده

We develop an analytical framework that allows central banks to assess whether changing the manufacturing material of their tokens would be beneficial. Applied to the case of the U.S., we find that a complete adoption of plastic notes would save the Fed $140 million per year but would entail a substantial migration cost in case of a “big bang”. On the level of individual denominations, we find that the $1 bill would be the most lucrative to replace and that the business case for the $100 bill is thin – suggesting that a partial adoption of polymer would make more sense.

Today, more than 30 central banks around the world have partially or fully adopted “plastic” (i.e., polymer) banknotes. The Bank of Canada (BoC) is a recent convert. The BoC issued its first plastic note, the C$100 bill, in November 2011 and by the end of 2013 all of its notes will be printed on the new material ( Bank of Canada, 2012). The primary driver behind the BoC's switch to polymer was the threat of counterfeiting. But another important advantage of plastic notes is that their lifespan is substantially longer than that of their paper counterparts. This yields considerable savings in ordering, processing, re-distribution, withdrawal and destruction costs. At least that is what Coventry (2001a) claims for the case of Australia – the country that was the first to successfully introduce a polymer banknote (in January 1988). However, these two advantages – improved protection against counterfeiting and higher durability – come with a downside: plastic notes are more costly to produce. As a result, not all central banks are convinced that the bottom line is positive. The Banco Central do Brasil (BCB) is a case in point. In 2000, it launched a trial with a commemorative note of 10 real. In the end, the BCB concluded that there was “insufficient evidence” of the benefits of polymer, and withdrew the notes ( Sidney, 2009, p. 11). For a central bank, calculating the gains of the adoption of a new banknote technology is indeed not that simple. In this paper, we therefore develop a formal framework that allows central banks to map both the cost drivers and the revenue streams that are affected. Our model is general in two senses. First, it can handle all relevant dimensions of the problem. Second, although we focus on banknote printing technologies when illustrating our model, it holds equally well for the replacement of notes by coins, or even for a change in minting technology. In another contribution, we use our model to estimate the cost savings brought about by the (complete or partial) introduction of plastic banknotes by the U.S. Federal Reserve, the Fed being a rare example of a central bank that releases information on the production costs of its notes. We show that while a complete adoption of plastic notes would entail a drop in monetary seigniorage revenue of roughly 0.2% (because of the higher initial production costs), it would cut by half the annual replacement cost of banknotes, resulting in net savings of $138.4 million per year. These yearly operating gains are to be compared with a one-off migration cost of as much as $3.3 billion in case of a “big bang” (but substantially less in case of a gradual migration). The remainder of the paper is structured as follows. In Section 2, we first explain how a central bank, by improving the cost effectiveness of its banknote production, can in fact make a contribution to solving the broader problem that is the social cost of cash. In Section 3, we present our theoretical framework. Section 4 is the empirical part of the paper. We describe the data of our case study and simulate how the costs and revenues of the Fed would be impacted by a switch to polymer. We also check whether one of the insights put forward in Menzies (2004) – the only other academic paper on the issue – is borne out by our analysis, as this enables us to present our results from a different perspective. Section 5 concludes.

This paper proposes an original framework that allows central banks to assess the operating gains of a change in the technology behind their monetary tokens. The model can assimilate all essential value drivers; that is, all sorts of costs, seigniorage, and even the impact on counterfeiting. Unlike Menzies (2004), our model is single-year instead of multi-period. This is, however, not a weakness. Our model could easily be made multi-period, but the value added would be limited, as each year is simply a repetition of the previous. Also, our model in fact handles currency growth better than Menzies’. This is not to say that there is no room for improvement. For example, we work with constant unit production costs, cp(j), and thus disregard economies of scale, whereas the attractiveness of plastic banknotes is clearly affected if the fixed costs, ceteris paribus, outweigh those of paper, and if currency demand were to drop. When applying our framework to the case of the Federal Reserve, we find that a complete adoption of polymer would entail a drop in seigniorage revenue of roughly 0.2%, but would cut by half the annual replacement cost, resulting in net operating savings of $138.4 million per year – not counting the recurrent gains from the decrease in counterfeiting (which are negligible).15 To put this figure in perspective, while substantial from the point of view of the Fed – the currency production cost reported in its Annual Report for 2008 amounts to $500 million – it only represents 0.001% of the average U.S. GDP over 2004–2008. If one compares this with the 0.04% of GDP that the Hungarian central bank puts forwards as the potential savings for society from a (drastic) switch from cash to cards 16, it becomes plainly clear that the strategy is effectively second-best, as pointed out in Section 2; see, however, below. On the level of individual denominations, we find that the $1 note would generate the biggest cost saving. More generally, the break-even horizon lengthens with the face value of the denominations (for our benchmark scenario the correlation coefficient amounts to 0.99). This is in line with the observation by Galán and Sarmiento (2007, p. 12) that in real life central banks adopt polymer mainly for their low-value denominations. The estimated magnitude of the potential operating gains for the Fed raises the question why it has not already switched to polymer. There are a number of possible reasons. For one, there is the one-off migration cost and the length of the associated break-even horizon – which would be no less than 35 and 41 years for the $50 and $100 bill, respectively. (The picture is even bleaker in terms of cumulative net benefits; see Table 4). Also, as Kristin Langwasser of the Reserve Bank of New Zealand pointed out to us, sometimes polymer notes are simply too durable: “a note lasting 20 years might not be needed if a central bank upgrades and replaces the entire series every ten years”.17 All this would argue in favor of a partial adoption of polymer. A second element is that central banks have made considerable investments in the development of special paper for their notes. The economic lifetime of such investments is quite long. A third element is specific to the U.S. and is related to the circumstance that over half of the value of U.S. currency is reportedly held outside the country (Judson, 2003). The U.S. Treasury might be worried that the introduction of notes with a radically different look and feel would cause foreign holders to think that the old notes are no longer “good”. This could trigger a sizeable “redemption run”, and thus a (minor) hit to the government budget. Finally, perhaps our benchmark scenario is too optimistic, especially where the lifespan of the polymer notes is concerned. Broadening the scope, and reasoning in the private/social-costs-framework introduced in Section 2, it could be argued that the switch to polymer may well lower the private costs of the central bank, but that it would seem to do precious little to lower costs elsewhere in the value chain – given that polymer notes remain a physical product and that the manual handling costs (including secure transport) thus do not disappear. Stronger still, the switch could even turn out negatively for other parties, in particular operators of ATMs, vending machines, transit systems, etc. For Canada, Spencer (2011, p. 6) estimates the conversion cost for these stakeholders at as much as C$75 to C$100 million (compared to annual savings for the BoC of C$25 million). But he is convinced that “the cash system will benefit in the long run”. Where New Zealand is concerned, Lang (2002, p. 55) points out that because the RBNZ retained the same size notes, the conversion of the ATM network required only a software adjustment. And the experience in Australia is that polymer notes are actually better for machine processing than paper notes: there are fewer jams, fewer service call outs, and maintenance staff can be reduced ( Coventry, 2001b). Crucially, Coventry also points out that “polymer notes have been a facilitator for further changes that may not otherwise have happened”. For example, because of their higher quality and durability, polymer notes allow increased recirculation between the various participants, leading to “a reduction in the socially wasteful excessive amount of churn and cross-shipping of notes”. This should help in recouping the conversion costs involved, and perhaps more than that. Coventry argues – but does not demonstrate – that “these efficiencies have reduced costs for the RBA and commercial banks thus adding to the overall benefits from the move to polymer” (emphasis added). Note that the central bank studies mentioned earlier all demonstrate that the own cash-related private costs of the central bank pale in comparison with those of other parties. In Norway, for example, the former account for a mere 1.2% of the total social cost of cash, versus 44.4% for commercial banks and 21.2% for subcontractors ( Gresvik and Haare, 2001, p. 20). The policy takeaway here is that when tinkering with the nature of its banknotes, a central bank should factor in the externalities on other stakeholders, so as to not only limit migration costs but also enable efficiency gains elsewhere in the payment chain. This is where the broader savings would appear to lie that could lift the social impact substantially above the 0.001% of GDP that we have identified. Finally, where the private costs of consumers are concerned, again it would seem that no research has yet been done. To sum up, it would be interesting to extend our approach to a calculation of the (net) social cost of plastic notes. But Coventry (2001b), for one, is convinced that the move to polymer has not only been “very good for the RBA” but has also – to phrase it in terms of the title of our paper – lowered the bill “for the community generally”.