Wednesday, December 7, 2016

Likely Economic Impact of Demonetization

On November 8, the Indian government announced that the two largest denominations of notes (Rs. 500 and 1000),would be discontinued immediately. These accounted for 86% of the Rs. 17.54 Trillion currency in circulation (amounting to Rs 15.08 Trillion in value).

It has been acknowledged as a bold move by many economists and citizens. The objectives of the policy were to root out black money, to discourage future generation of black money, and to root out counterfeit currency that was being used by terrorists.


I have been thinking about what could be the likely economic impact of demonetization. 


Demonetization is likely to have the following impacts:


1) Reduction of black money: This manifests itself in two ways:



  • New disclosures of unreported income and government seizures: So far Rs. 11.5 Trillion has been deposited (until December 6) in bank accounts. The Income Disclosure Scheme (IDS) which was operational before November 8 had yielded disclosures of more than Rs. 0.67 Trillion. Rs. 2000 crores of unreported income disclosures and Rs 130 crore of seizures have been reported from depositors after November 8 so far. The tax revenue to the government from this is likely to be Rs. 958 crores (as of Dec 7). This number is expected to rise as more disclosures and seizures are made. Future seizures of real estate, gold and other assets bought from unexplained sources of income is an area of great promise.
  • Old currency not deposited back in the banks. This leads to a reduction in the liability of the RBI (India's central bank) and could create an extraordinary profit for the RBI. Given that the deposits in the first 30 days since November 8 account for 76.3% of the value of old notes in circulation on November 8, many analysts and economists feel that less than 5% of the old stock of currency will remain in private ownership on December 31. The RBI Governor has clarified that old currency not returned back will not decrease the liability of the RBI, and therefore will not be recognized as income.
2) Printing cost for the new stock of currency (net of scrap value of old notes): According to the RBI , there were 16.5 billion ‘500-rupee’ notes and 6.7 billion ‘1000-rupee’ notes in circulation on November 8. Each 2000 Rs note costs Rs. 3.73 to print whereas a Rs. 500 note costs Rs. 2.97 to print. Due to the movement towards digital and plastic transactions, we could assume that the currency required to run the country's business could be reduced by 25-50% (Note that 26% of the deposits made till Dec 3 have already been withdrawn by citizens. That number is likely to rise as withdrawal limits ease and currency is made more available). My estimate is that the cost of printing an adequate stock of new currency would be between Rs. 3000 crores and 5000 crores


3) Decrease in value of housing and land stock: According to PropEquity research, (a real estate data and analytics platform covering over 83,650 projects of 22,202 developers across over 42 cities in India), valuation in the top 42 cities in India, sold and unsold, will take a tumble and fall up to 30 per cent ( approximately  Rs 8,02,874 crores. If we were to include older urban housing, rural housing and land, the impact will be several times more.


4) Impact on GDP: Demonetization is likely to effect national output in several ways:
  • Liquidity Reduction: As the availability of legal tender to conduct transactions is constrained for few weeks after the policy announcement, it impacts the ability of citizens and corporations in conducting transactions. This impact will get reduced only when adequate supply is made available and/or when citizens and corporations move to account-based transactions. As of October 2015, 233 Million citizens were still 'unbanked'. Even at full-capacity, the RBI printing presses will not be able to replace 50% of the stock of old currency in the next 4 months. This will impact cash-sensitive industries such as retail trade, hotels and restaurants and transportation, and in the unorganized (including agriculture) sector. These constitute collectively more than 50% of the economy, by my estimates.
  • Confidence Reduction: As liquidity got sucked out of the system in one fell swoop, consumers and purchasing managers become more frugal. Their confidence in spending their savings/income/profits gets impacted. This is visible in the latest Nikkei/Markit PMI Survey released in early DecemberThe Nikkei/Markit Services Purchasing Managers’ Index sank to 46.7 in November from October’s 54.5, the first time since June 2015 that the index has gone below the 50 mark that separates growth from contraction.It was also the biggest one-month drop since November 2008, just after the collapse of Lehman Brothers triggered the global financial crisis.“The latest set of gloomy PMI figures for the Indian service sector shows that companies were heavily impacted by the Rs500 and Rs1,000 banknotes ban,” said Pollyana De Lima, economist at survey compiler IHS Markit. The reduction in confidence will manifest itself across all sectors of the economy but particularly in luxury goods, automotive, jewelry, home appliances, and capital goods. These sectors have seen a 20-30% dip in business since November 8.
  • Activity Reduction in Construction: As real estate prices are expected to fall by 20-30%, construction of new homes will slow down. This will impact activity in the housing/ construction sector. As secondary transactions dry up in the remainder of the fiscal year, the registration taxes collected by state governments will also take a significant hit.
  • Productivity Loss: As people line up in queues for many minutes/hours per week (particularly in rural areas), there will be a significant loss in labour productivity. This impacts the unorganized and self-employed sectors more than other sectors.
  • Economic stimulation by increased government spending: The government can negate the impacts of the above effects by increasing spending. The Finance Minister, Mr. Jaitley has already proposed an increase in budget spending of Rs 35,170 crores to fund rural jobs development, the RBI Monetary Stabilization Scheme, farm subsidies and higher pensions. However this number is less than 1% of the quarterly GDP of India, and is unlikely to fully negate the other effects.
  • Vicious cycle effects: Even when liquidity is restored to 'acceptable levels' in 4 months, the ripple effects of the GDP de-growth will continue for a few months as people adjust to the 'new normal' (lost jobs/income in the unorganized sector, lost wealth for asset holders, etc). Of course, every government official and banker will try to convince the public that the future will be more rosy as the past, so that these vicious cycle effects are negated.
Various economists have indicated a 50 to 200 basis points reduction in GDP growth due to demonetization in FY'16-17. Due to the magnitude of the effects stated above, I personally believe that the impact will be much more. Even if the GDP for the 140 day period in FY'16-17 after the announcement reaches the same level as the GDP during the same period last year, the GDP growth for the year will have been impacted by more than 290 basis points. In my mind, it is likely that GDP impact is more than the gloomiest analyst's estimate (Ambit Capital's estimate of 330 bps impact in FY'16-17). Even if we assume that GDP growth is impacted by 330 bps in FY'16-17, that impact would total Rs. 4,95,000 crores!! Considering that taxes in India amount to 17.7% of GDP on average, the government's revenues for the last 2 quarters of the year will take a significant incremental loss.


Other Impacts: While I considered the impact of demonetization on future corruption and counterfeit currency, I believe these impacts will be short-term and insignificant. Counterfeit currency in circulation is estimated to be less than Rs. 400 crores. Counterfeiting will continue to be a challenge as counterfeiters find ways to duplicate the new currency.

What should the government do now?
  1. Increase government spending in the rural and unorganized sectors. The 37,000 crore relief being planned may not be enough!
  2. Increase social and welfare scheme funding to protect the weaker sections of society
  3. Aggressively mine data from bank depositors to identify tax cheats. Raid premises to seize gold/jewelry/property documents that have not been disclosed in tax statements
  4. Ensure that there is a taxable entity (PAN number) associated with each piece of real estate. Follow up with step 3.
  5. State governments should reduce property registration fees and property tax to prop up the secondary market, and to disincentivize creation of future black money.

(Disclaimer: Please note that I am a business analyst and not an economist. So please do take my analysis with the requisite amount of caution. It is definitely not an advice to buy or sell any assets.)


Thursday, January 30, 2014

In the 'Big Data' rush, let's not forget our 'Large Data' problems

As the Big Data hype cycle has swelled over the past two years, all sorts of people (hardware vendors, marketers, software vendors, business analysts, social media analysts, data scientists, etc) have jumped on to the bandwagon. Everyone wants a piece of Big Data on their resume, it seems.

Of course Big Data solutions (involving a NoSQL database and analysis to solve business problems that involve high velocity, variety, and volume of data) do have a huge potential to revolutionize marketing, customer care and supply chain in a host of industries.

However it seems to me that in this rush to find and solve problems involving 'Big Data', many companies are overlooking the potential to use large structured data-sets to transform their business. One of the applications of 'Large Data' is in real-time marketing. By 'real-time marketing', I am referring to the ability of the marketer to customize and automate marketing campaigns on a real-time basis.

If you think this is a trivial issue, answer the following questions:

1) How many marketers (OK, let's count Amazon, Macy's and Tesco out) send offers that are personalized to the browsing and transactional activity of each customer?
2) How many marketers put in an effort to customize their website landing page to the prior purchases of the customer or to the search term the visitor came from?
3) How many marketers customize the banners on their website to the preferences and the search history of the customer?
4) How many marketers follow the rule of not sending email to someone who hasn't responded to the last X offers? or set a rule for not sending more than Y communications to any particular customer?

...As it turns out, not a lot!

So as I said before, please get your 'Large Data' real-time marketing in order before you jump on to the Big Data bandwagon.