Demonetisation Lessons from Brazil

An edited version of this article appeared in Financial Express today. Link: http://www.financialexpress.com/opinion/note-ban-lesson-from-brazil-best-way-to-demonetise-is-not-to-have-one/472432/

Public policies are best when a lot of reason goes into their formulation and passion into their implementation.Those looking for an effective recipe for formulation could learn a lot from Brazil. It has demonetised its currency 8 times since 1942 and thrice simply knocked off the last 3 digits of its currency overnight i.e. like a 10,000 Cruzeiro (then Brazilian currency) will be 10 Cruzeiro from next day morning.

Lessons from 1830s to 1942.

Even before from 1830s it has been compelled to experiment with its currency due to evolving politics. The early experiments are to do with metallic convertible bases like silver and gold, metallic copper coins, birth of parallel paper money,  etc.

In early 1830s in order to stabilise the external value of Mil-Reis (then currency), the centre starved supply of currencies reducing the circulation of copper coins in the provinces. The provinces responded by issuing their own notes to neutralise demonetisation. Promissory Notes issued by Commercial banks valid for 15 days by law began to be accepted far beyond their due dates. (Source: Page 39-43,  Monetary Statecraft in Brazil: 1808–2014, Kurt Mettenheim)

Some other time commercial banks were allowed to issue bank notes (like in Hong Kong where currencies were issued by Standard Chartered and HSBC till accession). This led to loss of control of central authority and dilution of monetary policies.

Brazil through its history has clearly proved that no one can ‘starve’ the people of currency for far too long.

1942-1994

This period was mostly about high government expenditure, unbridled fiscal gaps and high inflation. Brazil demonetised 8 times before the last one in 1994.

It has had to change its currency, the ultimate form of demonetization for every conceivable reason – to tackle black money (Indian objective), to tackle hyper inflation, tackle daily cumulating interest rates of 3% (which is nearly 50,000% p.a.), base erosion, commodity price volatilities especially in Copper or just to avoid confusion (if Brazil had retained its currency same as in 1942, it would be 1 US $ =  2750 followed by 18 zeros, a nightmare for the accountants). They have been far deeper than t he Indian type demonetisation – the entire spectrum was replaced and the currency itself renamed.

The last in 1994.

The most recent in 1994 seemed Quixotic. It was aimed more at breaking the psychology of inflation. With 100% inflation consistently for 14 preceding  years (in 4 years over 1000%), shops had to revise prices 3 times everyday. That is when the government decided to use two currencies simultaneously – one virtual for counting the real value of currency and another for payments and settlement – and every shop having to display its prices in both and revise it 3 times a day.

But unexpectedly, people started anchoring their values against the real value (which was set near 1 Real Value unit = 1 US$).  Within a quarter or so, it was clear people were not rushing any longer to shops to avoid their currency buying less than when they started from home. Inflation abated and the real value became the Real the official unit. It was perhaps one of its most successful experiments that has lasted till date.

Lessons from Brazil

People will seek ways to settle transactions in the most cost and effort efficient ways. For many transactions in much of India, using currencies across the counter is still the most efficient option. In 1970s and 80s, when there was a coin shortage of sorts,  Chintamani co-operative superstore in Coimbatore used to issue their own tokens. These slowly gained acceptance with public so much so that even government owned busses and offices used them.

The parallel systems will start issuing notes and IOUs which will be strictly ‘enforced’ amongst its members through extra legal authorities.

One thing Brazil has always got right (between 1942-1994) is to have the 1,2,5,10,20,50,100 note sequence – considered the most friendly from transaction settlement point of view.

Currencies are as much about psychology and convenience as values for accounting and transaction, as the 1994 experiment so decisively proved.

The best way to demonetise is not to have one – avoid inflation, avoid unjustifiable or un-implementable tax systems, and not to issue too much of it anyway. Brazil has about 3% as currency/GDP whereas India’s is11-12%. Government should have incentivised and reduced it by 1% every year rather than force it in one lump.

A parade of demonetisations has not exactly curbed either parallel economy or corruption in Brazil. Corruption and black money is so rampant, their President was recently impeached for corruption, their biggest real estate tycoon is behind bars and may have to spend the rest of life there if not politically rescued.

Why black money or parallel economy, there is a near parallel administration being run by the mafia through drugs, extortion, violent thefts (one murder every 10 minutes i.e 140 a day, down of course from 600 a day not so long ago), etc. none of which will be happening through tax paid cheque money transfers.

Conclusion

In summary Brazil offers 3 ground rules (perhaps not with successful examples as much as negative narratives):

  • the way to tame inflation is not periodic demonetisations but curb state populism,
  • the way to curb black money and illegal economy is not starving people of cash but well thought out tax policies and effective punishments, and
  • the way to protect free trade from causing domestic unemployment problems is to maintain the external value of the currency which in turn is achieved by restricting external capital inflows to just what is required for financing current account deficits. (Donald V Coes, Macro Economic Policies and Growth in Brazil, 1964-90)

One would definitely give credit to both the government and RBI for curbing state populism within FRBMs. But given the levels of corruption in tax collection systems itself, black money curbing through demonetisation seems an ill fitting solution. Unemployment is rampant and growing due perhaps to highly overvalued Rupee and extra terrestrial real interest rates.

The daily dose of RBI circulars does indicate that someone is extremely alert at the wheel but whether he knows the destination and if it will deliver enough gains for the pains people are experiencing, time alone will tell.

The writer is CFO and author of ‘Making Growth Happen in India’ (Sage Publications)

India is not ready for CPI Based inflation targeting – not yet

It has been close to two years since the Urjit Patel Committee Report set the CPI based inflation ‘Targetting’ as the primary axis of our monetary policy. There are murmurs now from both the Vice Chairman of Niti Aayog and the CEA. In both its key thrusts (i) abandoning the multiple indicator approach for inflation control and (ii) adopting CPI combined instead of WPI as the inflation to target, it is looking like a cricket umpire trying to control the football game in the adjacent ground. With the benefit of hindsight, it looks out of context and tautological in its key arguments and conclusions.

Inflation targeting

“Anchored inflation expectations will … provide the latitude to address other objectives without compromising price stability” (Para II.3). In 18 months RBI has not been able to even ‘anchor’ interest rate expectations – there is so much of debate before each meeting and annoyance after. To anchor inflation expectation by stabilizing inflation to provide a stable interest rate regime to create ideal conditions for investments to generate employment and growth … by the time this comes about it may be Saturday in the Economic Solomon Grundy’s life cycle.

‘High inflation expectations exhibit far greater stickiness than inflation (para 3.2).’ In the past 5-6 years most of inflation has come from agricultural commodities and fuel – segments least under RBI’s control. Our agricultural markets are the most ‘perfect’ markets and given Indians’ mindset of ‘bargain everything’, prices quickly readjust to imbalances. If the Onion prices ruled at Rs 80 one week and slid to Rs 20 the next 2 weeks it is inconceivable that anyone’s expectation will be guided by the Onion prices that prevailed 3 weeks ago. The reverse is also true. Trying to stabilize them through interest rates is a fruitless exercise.

The level of emphasis to be given to inflation control should ideally depend on the national ‘wealth to current income’ ratio. Where it is high and more citizens depend on interest from savings for livelihood, preservation of money’s value is more important. India’s ratio will be dismal comparatively and hence there was a need to balance it with growth and employment objectives. In our context jobs are the best social security.

Constraining Targets

The targets set are a source of worry, for growth itself can cause inflation. For example, in a 2 product, 2 player economy producing 4 coconuts and 2 fish, each fish will retail for 2 coconuts. If the productivity of coconuts increases to 6, the price of fish will become 3. In a monetized economy, wages are terribly sticky downwards and the price of coconuts (derived from the wages which do not move downwards) will remain the same and the price of fish will move up by 50%. There will be the inevitable inflation even if the weights are corrected. This is harmless inflation however. Suppressing this will only result in curbing growth.

A constant inflation target of 6% (+/- 2%) irrespective of whether the growth is 4.5% or 8% seems meaningless.Given India’s rigidities and the way minimum wages are revised whimsically, a 6% target may be far too constraining for a 8% growth target.

In a place like Singapore where trade credits are near totally from banking system, and firms are leveraged 3 or 4 times and work on thin net margins of 2 percent, a ¼ % is a huge dampener and firms might start cutting down on stocks from the next cycle itself. In India with 10-12% gross retail margins a large proportion of which is imputed labour and imputed rent – both far less sensitive to interest rates – and credit largely accessed from non-banking sources, a ¼ % interest rate adjustment to tame or stoke inflation seems irrelevant. Large dosages to achieve a given reduction in overall inflation will hurt a whole host of other sectors disproportionately.

Pitfalls of a statistical approach

The report relies on the New Keynesian Philips Curve equation as the theoretical framework. The 3 factors listed (output gap, cost push, and expectations) in the supply block of equation while relevant is far from decisive or comprehensive. Let’s see an example – how ‘cost push’ can be highly episodic or fickle.

The supply curve is the marginal costs of various firms stacked in increasing order from most efficient to most inefficient and the price is determined purely by the marginal cost of the most inefficient firm required to fulfill a given level of demand. The cost structure of all the other more efficient firms stand irrelevant. Where the most marginal supplier unit (which determines the price) happens to be an overseas firm, prices will be purely determined by import parity. Domestic cost structures do not matter at all: what will matter is the cost structure of source countries. In such a case currency movements play a much larger role.

Again, where the marginal cost difference between the least competitive firm and the next is low or negligible, output gaps may not have an impact on inflation at all.

One does not know how one can form firm action-effect inflation targeting policies on such highly fickle variables alone.

Flawed assumption

The report justifies targeting since ‘persistent inflation worsens income distribution as the poor carry greater proportion of cash’(II.1). This is spurious sympathy. Poor carry cash largely for transaction demand. In rural areas savings till the next season is largely in grains and savings over longer term is in Gold. Surplus cash in the informal sector finds its way largely into small unorganized chit funds and informal credits earning 5-6% per month. It is impossible to conceive that they carry cash over longer time as savings. A cash balance of 2 months when inflation is 12% per annum suffers 2% value dilution. This is just 1% higher than at the targeted inflation levels – hardly relevant which only shows a lack of contextual knowledge.

Inflation in a way represents existence of consumer surplus. In the initial wave, it’s only products with high consumer surplus that will move up in price. There is no reason for RBI to be an arbiter in such a case. It is only in the secondary waves prices of others will move up before it becomes a monetary phenomenon. If those affected in secondary wave are interest rate sensitive then to target inflation with interest rates becomes logical: but only if.

Any regulator will have to keep the aspirations of the people in mind. Ask any job seeker whether he would prefer a job or live with an additional 5-6% inflation. Employment will most certainly be preferable to preservation of value. Growth expands employment opportunities. To be dismissive of these almost mockingly and concentrate on CPI based inflation is to forget the context.

One wishes the Committee had taken advantage of recent advances in Behavioral Economics while formulating its recommendations.

Link to the article published in Businessline: http://www.thehindubusinessline.com/opinion/inflation-targeting-makes-no-sense/article8253638.ece