The dual purpose of negative interest rates
and the role of cash currency
Bachelor thesis by Jan Lebens (10203958) Supervisor: Ioana Neamtu
Page | 2
Statement of Originality
This document is written by Student Jan Lebens, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Page | 3
Table of contents
Abstract 4
Introduction 5
Literature review 6
On negative interest rates and the zero lower bound 6
On abolishing currency 7
Empirical analysis 9
The zero lower bound on nominal interest rates 9
Negative interest rates in practice 11
The phasing out of cash currency 14
Discussion 16
Conclusion 17
Page | 4
I. Abstract
During times of financial crisis, policy makers have considered themselves incapacitated by the inability to set the interest rate below zero. They fear that if interest rates would be set below zero, consumers would withdraw their bank balances to avoid having to pay it. As such, cash currency is considered to be the primary burden to negative interest rate setting. That being said, the Swiss Central Bank has set its interest rate below zero in early 2015. It is the finding of this paper that the Swiss negative interest rate experience is not suitable for concluding that negative interest rates can be employed as a classic monetary policy instrument in times of crisis. It is due to the safe haven currency status of the Swiss Franc that the policy is not undermined by
substitution. To enable negative interest rate setting in order to stimulate expenditure, one solution would be to eliminate cash currency to some degree, which would also lead to other positive externalities.
Page | 5
II. Introduction
When the Swiss Central Bank became the first central bank to impose a negative interest rate on excess reserves in 2011, the existence of a zero lower bound on interest rates could be put to question. The zero lower bound problem, which had enjoyed little academic interest since the Great Depression of the 1930’s, had resurfaced in the context of the Japanese crisis of the 1990’s. It was during this crisis that interest rates set by the Central Bank of Japan came closest to the lower bound, which is considered to be at zero percent. When the developed economies nearly all came to face the same problem during the Great Recession that started in 2007, the ability to set interest rates below zero became an option that some considered useful to speeding up economic recovery. The reason that there is thought to be a lower bound in the first place being that deposit holders will simply substitute their holdings into cash currency or other liquid assets, thereby avoiding the interest payments. The purpose of this thesis is to assess whether the case of Switzerland can be considered evidence that the zero lower bound is a concept and not a practical limitation, and whether the abolishment of cash currency can be expected to enable policy makers to set the interest rate below zero at will. The findings will determine how proponents and opponents of negative interest rate setting will be able to use the case of Switzerland as an example. It will also contribute some insight as to the role that cash currency has in this discussion.
In order to be able to formulate an answer to the research question, I will provide the results of my study in three parts. In the first section I will focus on the relevance of the problem and its possible solutions. Here I will provide insight as to why a zero lower bound on interest rates might be a problem that policy makers wish to overcome and what means are being proposed to do so. This section will be a combination of literature review and observations, as I will include results from other studies that have focused on the Lost Decade of Japan, during which Japan became the first of the developed economies to lower interest rates to near zero levels.
In the second part I will take a closer look at an example of negative interest rates in practice, specifically the case of Switzerland between early 2015 and end 2017. Following from the results of the first section, it is worth to analyze why there can be discussions on the zero lower bound on interest rates when there are central banks already implementing them such as the Swiss National Bank, Denmark’s Nationalbank and Central Bank of Sweden. Based on the data and reports provided by the Swiss National Bank, I will conclude in what way these policy decisions are relevant to the zero lower bound problem.
Lastly, I will return to the theoretical and discuss the pros and cons of phasing out paper currency, and why that might be necessary. Although the development of electronic payment means is
Page | 6
rapidly advancing, relative to the Lost Decade times, no country has abolished its paper currency as of yet. As such, this section will again take the form of a literature review.
III. Literature review
A. On the history of the zero lower bound
Over the history of the federal funds rate it has occurred in two periods that the rate approached the zero lower bound. The first of which was during the Great Depression in the 1930’s, the second the Great Recession of 2007. It was after the Great Depression that economists started to evaluate its properties and means to overcome it. As described by Ilgmann and Manner (2011), the first of such means proposed has been the taxation of held currency by Gesell (1916). The concept being that currency would be
periodically taxed to maintain its validity and to create a burden on hoarding. It formed the foundation for the Stamp Scrip by Fisher (1933), and was even put to use in small scale experiments (e.g. Miracle of Wörgl). As the crisis faded however, so did the academic interest for the zero lower bound on interest rates. It wasn’t until Japan faced its Lost Decade the 1990’s that another developed country experienced similar conditions, as pointed out by Krugman (1998). In his paper, Krugman describes the zero lower bound to be a property of the liquidity trap. The liquidity trap being the situation in which further expansion of the monetary base under zero interest rates yields no results due to the perfect substitutability of money and bonds.
B. On the proposals made to overcome the liquidity trap
Krugman (1998) writes that overcoming a liquidity trap will be a matter of managing expectations and credibility, as he concludes that interest rates can “clearly” not go negative. The reasoning offered for this is that money would then dominate bonds as an asset. Eggertsson (2003) states that an economy has fallen into a liquidity trap when “interest rates have fallen to a level below which they cannot be driven further by monetary expansion” in combination with a credibility problem that undermines any target
announcements. The credibility is to be managed by demonstrating resolve in the period leading up to the period in which the zero lower bound is reached (Eggertsson and Woodford, 2003). Together with the inability of monetary base changes to influence the inflation rate, the zero lower bound on interest rates is one of the two key features of the liquidity trap.
Page | 7
Another proposal to overcome the zero lower bound is a paper by Svensson (2000), who formulated his recommendations prior to the Japanese making the commitment to engage in quantitative easing. His recommendation is that the “foolproof” way of escaping a liquidity trap is the combined usage of a price-level target path, a devaluation of the currency and a temporary exchange rate peg. The aspect of
devaluation is agreed upon by Bernanke (2000). His recommendation however involves large open-market sales of Yen as a means to achieve devaluation. He states that, as Krugman (1998) also writes, the open-market sales must be credibly large and sustained in order to overcome the liquidity trap.
C. On the proposal for negative interest rates
Others do consider taxation as a solution to the Japanese deflation problem. These include Goodfriend (2000), Fukao (2005) and Buiter (2003, 2009). Although the aforementioned authors are not entirely in agreement on which financial assets should be taxed, all propose it as a means to overcome the zero lower bound problem. Goodfriend (2000) finds that the primary obstacle to the realization is the option of substitution into cash currency, and both he and Fukao discuss Stamp Scrip-like solutions to overcome it. Buiter (2009) has gone even further by proposing the abolishment of cash currency to prevent
substitution, which will be further highlighted in the latter part of this literature review.
Sumner (2002) offers a contradicting position, stating that a pure liquidity trap cannot occur as the central bank would never be able to inject unlimited amounts of currency into the system without affecting other variables. The counterargument offered by Krugman’s reasoning is that if the expansion is not expected to be lasting, the shrinkage that will follow will undo the initial effects.
D. On abolishing currency
Buiter (2009) suggests three possible measures to overcome the zero lower bound, one of which is the abolishment of cash currency. A means of payment that he argues is already becoming redundant in the developed economies due to the growing range of electronic means of payment. The usage of cash currency in these economies is decreasing to small retail payments and shadow economy transactions. Rogoff (2015; 2017) writes that the small retail payment function can be preserved by abolishing large denomination notes only, thereby making it more difficult to carry large amounts of cash as it would simply take up more volume and weight. Both Rogoff and Buiter discuss an intermediate solution in which cash currency is regarded as a separate currency, deposit balance and cash currency being
exchangeable. Thereby preserving its functionality as a means of payment in small retail transactions, but also enabling deposit balance taxation.
Page | 8
E. On the feasibility of cash abolishment
Svensson (2000) and Goodfriend (2000) also comment on the ability of a central bank to tax commercial-bank reserves but reckon it would require technological innovations such as electronically chipping notes. This form of technology might have been seen as being advanced at the time of writing, but is an outdated concept presently. Rogoff (2015) for example already comments on the technological capabilities of the crypto currency markets and its potential for ultimately strengthening the electronic payment options. Agrawal and Kimball (2015) go as far as to propose a potential step by step guide to replace cash
currency with electronic currency. The ability to withdraw deposits from commercial banks does however does provide protection from “irresponsible” policy advise, argue Berentsen and Schär (2015). The decision to withdraw however may not be only motivated by “irresponsible” policy advice, as for example bank runs may occur for a variety of reasons. Abolishing cash currency makes it impossible for non-banks to withdraw funds from the banking system (Krueger, 2015).
Page | 9
IV. Empirical observations
The primary reason to consider abolishing cash currency to some degree is that it would enable monetary policy makers to pierce the lower bound on nominal interest rates. Interest rate setting is used as a monetary tool to regulate the incentive to save and borrow. In times of economic contraction, interest rates are generally lowered to stimulate borrowing, counteract the hoarding of risk-free liquidity and decrease the required return on assets. Conventionally, the interest rate is lowered to the point that the central bank believes is required. It is possible however that the determined equilibrium interest rate is below zero, turning savings into a guaranteed loss rather than a guaranteed return.
A. The zero lower bound on nominal interest rates
The first developed country to experience limited nominal interest rate setting, since the 1930’s, has been Japan, during its “Lost Decade” that span from 1991 to 2000. In that period, the Bank of Japan (hereafter: BoJ) has lowered its basic loan rate to below 1.00% for the first time in its existence. When it failed to reinvigorate the dwindling inflation rate, the BoJ claimed it had done all it could (Krugman, 1998). Although further lowering the basic loan rate to 0.00% after 2002 (see figure below), claiming it could do no more meant that the rate had reached its effective lower bound.
Figure 1. “Basic discount rate (BoJ)” Basic discount rate between 1985 and 2005. Source1 :http://www.stat-search.boj.or.jp/ssi/cgi-bin/famecgi2?cgi=$graphwnd_en
When the nominal interest rate is zero, quantitative easing can become ineffective due to bonds and cash essentially become perfect substitutes (Krugman, 1998). When the situation arises in which this is true, the economy is in a liquidity trap. One element of which is the assumed to be binding zero lower bound.
Page | 10
Some consider the zero lower bound to be a genuine constraint, and argue this does not mean that equilibriums cannot be achieved through some other means than interest rate setting. Instead, central banks can resort to solutions such as the management of expectations regarding interest rates, price-level targeting, currency devaluation and inflation targeting (Krugman, 1998; Bernanke, 2000; Svensson, 2001; Eggertsson, 2003). The issue with such proposals is that if the central bank’s credibility is lacking, it will undermine its ability to influence expectations.
The Financial Times article (Guha, 2009), quoted in both Buiter (2009) and Ilgmann & Menner (2011), refers to an internal report by Federal Reserve staff that projects the ideal federal funds rate at negative five percent towards the end of 2010 (Greenbook part 1, March 12, 2009). The line that reaches and stays at 0%, labeled “Constrained monetary policy”, indicates the perceived existence of a zero lower bound on nominal interest rates. The reason for the expected effect on the employment rate being that increased expenditure, as a result of the negative interest rate, will lead to job creation.
Figures 2 and 3. From the Greenbook part 1, March 12, 2009.
The constraint is “perceived” as it is only due to the decision of the policy makers that the interest rate did not pass below zero. A New York Times article by Mankiw (2009) also considers the setting of negative interest rates, but acknowledges that consumers might simply withdraw their deposits and hold their balances in cash. This is considered to be the primary obstacle, by both Rogoff (2017) and Buiter (2009), to negative interest rate setting. The reasoning that the interest rate cannot be below zero (Krugman, 1998) is especially interesting as several countries, including Switzerland and Sweden among others, set their rates below zero after the Great Recession that started in 2007.
Page | 11
If the reasoning that negative interest rates cause substitution into bonds, cash currency and other non-negative rate bearing assets upholds, then I would expect to see a corresponding decrease in commercial bank reserves held at the central bank. In the following section the case of Switzerland will be examined more closely to this end.
B. Negative interest rates in practice
Observations of interest rate setting in practice can lead one to conclude that a binding zero lower bound is in fact a myth (Jarrow, 2013). With commercial banks and central banks setting rates below the zero, it is hard to claim otherwise. An example of a central bank that has done so is the Swiss National Bank.
Figure 4. Swiss National Bank interest rate (right axis) and sight deposit balances (left axis) between 1999 and 2017. Source2: https://data.snb.ch/en/topics/snb#!/cube/snbbipo
The Swiss National Bank (hereafter: SNB) introduced negative interest rates on reserves in January 2015. Although a clear trend can be observed in the figure above, from the end of 2007 onward, the change in interest rates seem to have no apparent effects on the growth of the SNB sight deposit balance. In fact, the sight deposit balance with the SNB has grown in spite of negative interest rates. The observation that this opposes what has come to be expected of negative interest rates, substitution into cash currency, I argue, can be explained in four ways. Firstly, the interest rate is only negative on reserves that are in excess of twentyfold the minimum requirement. This could lead one to conclude that there were not enough assets exposed to the negative rates to have any impact. Given however that the reserves exceeded the
requirement by an annual average of 420.8 billion Swiss Francs (hereafter: CHF) in 2016 (SNB Annual Report 2016), and negative interest rate income totalled at CHF 1.5 billion that year, a large share of the 2: Retrieved on 07/12/2017.
Page | 12
reserves is affected. Institutions that maintain a balance with the SNB, but are not required to have reserves, are required to pay interest above an undisclosed threshold. Given that 86.75% of sight deposits is held by domestic banks (as of November 2017, SNB), this will not be further looked into.
Secondly, the purpose of the negative interest rate, as stated by the SNB, is to make the CHF less attractive and thus ease upward pressure on the currency. Danthine (2017) concludes that the goal of restoring a desired interest rate differential for safe haven currency countries is achievable using negative interest rates, and that the policy can even be strengthened with reasonable efforts. He doubts that the policy would be able to provide the “classic” economic stimulus by observing that bank lending rates barely changed as a result of the rate change and therefor very little macroeconomic stimulus is to be expected from the policy.
Thirdly, a possible explanation is that the inflation rate in the same period was such that the real interest rate was still positive. I have calculated the real interest rate, using the Fisher equation (nominal interest rate being approximately equal to the sum of the real interest rate and the inflation rate), reveals that the real interest rate was positive after the introduction of the negative interest rate on sight deposits, but has been declining due to the rising inflation from mid-2015 onwards.
Figure 5. Swiss inflation, nominal- and real interest rate between 1999 and 2017. Source3: https://data.snb.ch/en/topics/ziredev#!/cube/zimoma
I argue that positive real interest rates might counteract the expectation of substitution is the following. If a deposit holder expects the relative value of his deposit, measured in purchasing power of consumer goods, to increase by more than the interest payment, he or she will not be incentivized to substitute into
Page | 13
non-interest bearing assets. The observation that the inflation rate increased following the negative interest rate setting could be interpreted as a result and evidence that the policy measure did bring about increased expenditure. Contributing the increase of the inflation rate entirely to the negative interest rates is not straightforward, as euro area inflation rates performed similarly over the same period, suggesting systematic causes, as displayed in the figure below.
Figure 6. Euro area and Swiss inflation between 1999 and 2017. Source4 of Swiss inflation rate as in the figure above, source5 of
Euro area inflation rate: http://sdw.ecb.europa.eu/
Lastly, another explanation as to why no substitution into cash currency, or alternative, occurred is that interest rates were not negative enough to take on the risk of carrying large amounts of cash currency. Due to the size of the volumes of currency deposited at the central bank, it is more unlikely for
institutions to withdraw into than small deposit holding consumers. Holding cash entails certain costs that electronic deposits do not create (Rogoff, 2016). If future economic crises demand an interest rate lower than any observed to date, I expect the likelihood of substitution to grow as the interest rate drops. As such, the case of Switzerland is not conclusive proof that the zero lower bound is not a constraint in times of financial crisis. To at least prevent substitution into cash currency, measures can be considered that are outlined in the following section.
Page | 14
C. The phasing out of cash currency
As can be concluded from the previous sections, the cost of holding cash currency is large enough to enable policy makers to set below zero interest rates, while maintaining the usage of cash currency. Although there is likely to be a below zero lower bound at which the substitution does occur, there are other externalities of abolishing cash currency to consider.
One externality of abolishing cash for the purpose of enabling negative interest rates is that cash currency can no longer be used as a means to evade taxation (Berentsen & Schär, 2015; Rogoff, 2015, 2017; Buiter, 2009). Since wealth tax calculated based on the balances value date the first of January each year (Netherlands, assuming the same for Switzerland), and cash currency holdings cannot be readily
examined by the tax authorities, tax evasive behavior would indicate an increase of currency in circulation in December each year. Continuing with Switzerland for this example, the below graph displays the cash currency in circulation, as a sum of all notes and coins, and the relative retail turnover between 2002 and 2017;
Figure 6. “Cash currency and retail turnover”, major vertical gridlines indicate year-ends. Source6 of cash currency in circulation data: https://data.snb.ch/en/topics/snb#!/cube/snbmoba, source of retail turnover data: https://data.snb.ch/en/topics/uvo#!/cube/conretail
The small spikes in the volume in circulation are ranging from 4.51% increase in December 2016 to an 11.65% increase in December 2008. The volume then decreases again in January by 8.26% and 2.33% in the same years respectively. At its height, the currency in circulation per capita reached CHF 9,294.52 (2016 population estimate of 8,401,120) or EUR 7,926.56 (as of 24/01/2018, at 1.17258 CHF/EUR from
Page | 15
www.xe.com). The retail turnover, in percentage points, is relative to the 2010 level. Although the increase in cash currency in circulation towards the yearend could be interpreted as substitution into cash for tax evasive purposes, the spikes coincide with periodic increases in retail turnover. The increase in retail turnover is arguably due to the holiday period. The period-on-period change is characterized by a correlation of 0.783, but it is unclear which share of the increase in retail turnover paid for in cash; as such data is difficult to gather due to the anonymous nature of cash currency. As such, abolishment of cash currency would provide policy makers full information with regard to the currency flows in the economy.
A second externality of the abolishment of cash is the burden imposed on shadow economies, to which tax evasion is also considered to belong. Buiter (2009) argues that the only beneficiaries from the existence of anonymity-providing currency are the underground economy. For the case of Switzerland, it is estimated that the shadow economy was equal in size to 6.9% of gross domestic product in 2014 (Schneider et. al, 2015). Switzerland is estimated to have the smallest relative shadow economy of all European countries, with Bulgaria being estimated at 31.0% and the European non-weighted average at 22.1%. Although abolishing cash is not expected to directly decrease crime rates, consider the following example inspired by Rogoff (2016); taking the large denomination notes out of circulation means that a briefcase of CHF 1,000,000 in CHF 1,000 notes weighing 1 kg, will weigh 100 kg in CHF 10 notes, if that becomes the largest denomination. In 2016, 62.1% of the value of all cash currency in circulation comprised of CHF 1,000 notes (10.6% of notes) (SNB). As such, abolishing the larger denomination notes will create an increased burden to carry, verify, handle and store (Rogoff, 2016).
Given these positive externalities, adding the resolution of the zero lower bound problem, it is unclear why cash currency still has a role in our society today, when electronic payment options are at the current stage of development. One argument brought forward by Buiter (2009) and Rogoff (2015) is that the central bank generates seigniorage revenues by printing coins and notes. In the case of the SNB, profits are distributed to the Confederation and cantons. Including the revenue made charging negative interest, the total distribution made in 2015 and 2016 was CHF 3 billion. Considering that an increase in tax revenues is to be expected from the riddance of at least the larger denomination bank notes, the argument of the loss of seigniorage revenues is negligible in my opinion.
Page | 16
V. Discussion
The analysis outlined above is focused on only one solution to the zero lower bound problem, the abolishment of cash currency as proposed by for example Buiter (2009), alongside two other possible solutions. Although there are externalities that favor the abolishment of at least the larger denomination notes in my opinion, similar results could also be achieved through means such as cash withdrawal limitation and restricting the size in which cash payments can be made. Such measures will hinder the expected deposit outflow that occurs as a result of the incentive to avoid having to pay the interest. Substitution into other liquid assets than cash currency is also possibility worth considering. As briefly noted, highly liquid crypto currencies that are becoming more and more accessible to the general public may undermine any benefits that have been outlined above.
In the analysis of the case of Switzerland, only commercial bank reserves with the central bank are considered as assets exposed to negative interest rates. In practice, I expect negative deposit rates to be more likely to generate an effect on smaller consumer balances, as they are more easily withdrawn than commercial bank reserves. Holding cash brings about costs that electronic deposits do not (Rogoff, 2016), such as protection against theft and possibly storage given that the amount is large enough. The smaller the amount of cash held, the smaller the costs imposed on the holder, and as such consumers can be expected to be more prone to withdraw their deposits should they be forced to pay for them. As long as substitution into cash currency is an option, commercial banks may be hesitant to impose a negative interest rate as it would negatively affect its deposit base. The limited transmission of interest rates because of this potentially hinders the effect of the policy measure as a result (Arteta et al., 2016). Provided that commercial bank lending rates were hardly affected in the case of Switzerland (Danthine, 2017), no current example of negative rates can provide evidence of substitution, if the Danish Krone and the Swedish Krona are also considered to be safe haven currencies.
Page | 17
VI. Conclusion
The abolishment of cash currency can provide means to overcome the zero lower bound on nominal interest rates, by creating a burden to substitution into cash currency as a way to evade having to pay interest. The degree to which cash currency needs to be abolished however, is likely to be limited to larger denomination bank notes rather than all forms of cash currently in circulation. This can be expected to bring about other positive externalities such as reduced tax evasion, increased tax revenues and an increased burden on shadow economy transactions and capital accumulation.
Negative interest rates can be employed for two purposes, one of which being the creation of a desirable interest rate differential thereby influencing the exchange rate to counter unwanted domestic currency appreciation. The other purpose is considered the classic purpose, which is the creation of a burden on hoarding savings in times of economic contraction, to stimulate expenditure. The case of Switzerland, and thereby arguably any safe haven currency country, is not fit for evaluation of the
negative interest rate as a monetary tool used for its classic purpose. Currently, there is no evidence of any central bank setting its interest rate at a level such as suggested to the Federal Bank in its internal report. Therefore, the zero lower bound is, although practically broken, still relevant for interest rate setting for the classic purpose.
Page | 18
VII. Reference list
1. Agarwal, R. & Kimball, M., 2015, “Breaking Through the Zero Lower Bound”, IMF Working Paper 15/224.
2. Arteta, C., Kose, A., Stocker, M., & Taskin, T. (2016). Negative interest rate policies: Sources and implications.
3. Berentsen, A., & Schär, F. (2015). The fallacy of a cashless society. Cash on Trial, 14. 4. Bernanke, B. S. (2000). Japanese monetary policy: a case of self-induced paralysis?. Japan’s
financial crisis and its parallels to US experience, 149-166.
5. Buiter, W. H., & Panigirtzoglou, N. (2003). Overcoming the zero bound on nominal interest rates with negative interest on currency: Gesell's solution. The Economic Journal, 113(490), 723-746. 6. Buiter, W. H. (2009). Negative nominal interest rates: Three ways to overcome the zero lower
bound. The North American Journal of Economics and Finance, 20(3), 213-238. 7. Danthine, J. P. (2017). Negative interest rates in Switzerland: What have we learned?.
8. Eggertsson, G. B. (2003). Zero bound on interest rates and optimal monetary policy. Brookings papers on economic activity, 2003(1), 139-233.
9. Eggertsson, G. B., & Woodford, M. (2003). Optimal monetary policy in a liquidity trap (No. w9968). National Bureau of Economic Research.
10. Financial Times (2009). Fed study puts ideal interest rate at -5%, Guha Krishna, April 29. 11. Fisher, I., Cohrssen, H. R., & Fisher, H. W. (1933). Stamp scrip. New York: Adelphi Company. 12. Fukao, M. (2005). The effects of ‘Gesell’(Currency) taxes in promoting Japan's economic
recovery. International Economics and Economic Policy, 2(2-3), 173-188. 13. Gesell, S. (1958). The natural economic order. Owen. (Revised edition)
14. Goodfriend, M. (2000). Overcoming the zero bound on interest rate policy. Journal of Money, Credit and Banking, 1007-1035.
15. Ilgmann, C., & Menner, M. (2011). Negative nominal interest rates: history and current proposals. International Economics and Economic Policy, 8(4), 383-405.
16. Jarrow, R. A. (2013). The zero-lower bound on interest rates: Myth or reality?. Finance Research Letters, 10(4), 151-156.
17. Krueger, M. (2016). Pros and cons of cash: the state of the debate. Cash on Trial, 45. 18. Krugman, P. R., Dominquez, K. M., & Rogoff, K. (1998). It's baaack: Japan's slump and the
return of the liquidity trap. Brookings Papers on Economic Activity, 1998(2), 137-205. 19. Mankiw, N. G. (2009). It may be time for the fed to go negative. New York Times, 18, 2009. 20. Rogoff, K. (2015). Costs and benefits to phasing out paper currency. NBER Macroeconomics
Page | 19
21. Rogoff, K. S. (2016). The Sinister Side of Cash. The Wall Street Journal, 25. 22. Rogoff, K. (2016). The curse of cash. Princeton.
23. Rogoff, K. (2017). Dealing with Monetary Paralysis at the Zero Bound. Journal of Economic Perspectives, 31(3), 47-66.
24. Schneider, F., Raczkowski, K., & Mróz, B. (2015). Shadow economy and tax evasion in the EU. Journal of Money Laundering Control, 18(1), 34-51.
25. Sumner, S. (2001). some Observations on the Return of the Liquidity Trap. Cato J., 21, 481. 26. Svensson, L. E. (2000). The zero bound in an open economy: A foolproof way of escaping from a
