A bad bank for bad banks?

RBI-Logo_8

India’s public sector banks(PSBs) are in a big mess. As of September 30, 2016, the gross non-performing assets ratio of these banks stood at around 12 per cent. A gross non-performing assets ratio or a bad loans ratio of 12 per cent basically means that for every Rs 100 loaned out by the banks, the borrowers have
stopped paying interest on Rs 12.

One solution that has been advocated to solve this problem is that of a bad bank. To put it simplistically, the solution entails moving the bad loans of the public-sector banks to a newly created bank. This bank, referred to as the bad bank, will then go around recovering the loans that have been defaulted on by selling the assets offered as a collateral against the defaulted loans.

The PSBs will have to be recapitalised by the government and then they can simply concentrate on the lending business. The bad-bank strategy was successfully followed in the United States to sort out the Savings and Loans crisis of the 1980s. It was also used successfully in Sweden in the early 1990s.

The latest Economic Survey released on January 31, 2017, talks about setting up of the Public Sector Asset Rehabilitation Agency(PARA). This will be a bad bank which will buy the bad loans from the PSBs and “then work them out, either by converting debt to equity and selling the stakes in auctions or by granting debt reduction, depending on professional assessments of the value-maximizing strategy”. The bad bank strategy has also been recently recommended by Viral Acharya, a deputy governor of the Reserve Bank of India.

The thing is, this is not as simple as it sounds. In the past, PSBs have tried to sell their bad loans to private asset reconstruction companies. Like in case of bad banks, a bank sells its bad loans to an asset reconstruction company, which then goes about selling the assets held as collateral against bad loans.

The trouble is that the asset reconstruction companies haven’t really been able to do a good job of it. As the Economic Survey puts it: “Asset reconstruction companies have found it difficult to resolve the assets they have purchased, so they are only willing to purchase loans at low prices. As a result, banks have been unwilling to sell them loans on a large scale.

This is a problem that a bad bank will also face. At what price should it buy the bad loans from the public sector banks? Will those banks be ready to sell at that price, given the fear of courts, vigilance as well as the CAG?

Assuming the banks and the bad banks are able to get over this obstacle, they will run into another major obstacle. Many corporates to which banks have lent money have an interest coverage ratio of less than one. These companies are referred to as stressed companies in the Economic Survey. This basically means that the operating profit (earnings before interest and taxes) of these firms is lower than the interest that they need to pay on their outstanding debt, during a given period. They are simply not earning enough to be able to pay the interest that is due on their debt.

The stressed companies with an interest coverage ratio of less than one, owe a little more than 40 per cent of the loans given out by Indian banks. In fact, even within stressed companies the problem is concentrated among a few borrowers. A mere 50 companies account for 71 per cent of the loans owed by the stressed companies. On an average these companies owe Rs 20,000 crore each to the banking system. The top 10 companies on an average owe Rs 40,000 crore apiece.

These are some of the biggest business groups in the country. Going about selling their assets in order to recover the loans will not be easy for the bad bank. These groups have access to some of the best legal brains in the country. They are also close to the politicians. As Acharya put it: “I don’t think a bad bank just by itself will necessarily work, I think it has to be designed right.”

Political will allowing the bad bank to go after business groups which have defaulted on bank loans, must be a big part of that design. Does the Modi government have that will, is a question worth asking?

(The article was originally published in the Daily News and Analysis(DNA) on February 16, 2017).

All Bank Deposits Post Notebandi Have Been Invested in Govt Securities

rs-500

One of the many theories offered in favour of demonetisation or notebandi, has been that it has led to lower interest rates. The demonetised notes of Rs 500 and Rs 1,000 had to be deposited into the banks, where the money would be credited against the depositors’ name.

This has essentially led to the total amount of deposits with banks increasing at a rapid rate. Between October 28, 2016, before the demonetisation happened, and January 20, 2017, the total deposits of banks went up by 5.7 per cent. This increase in a period of under three months is huge.

With an increase in deposits, banks have cut interest rates. The logic offered by many experts was that this cut in interest rates will lead to an increase in lending by banks and that would be good for the economy. But the data suggests that anything like that hasn’t happened.

Take a look at Figure 1. It essentially shows the portion of bank deposits that have been invested in government securities, in the recent past.

Figure 1:

 

What does Figure 1 tell us? As on October 28, 2016, before demonetisation was carried out, around 29 per cent of bank deposits had been invested in government securities. As on January 20, 2017, the latest data that is available, the proportion had jumped to 34.1 per cent. Banks need to compulsorily invest just 20.5 per cent of their deposits in government securities.

Further, between October 28, 2016, and January 20, 2017, the total amount of money banks got as deposits stands at Rs 5.63 lakh crore. During the same period, the total amount of money banks invested in government securities was Rs 6.99 lakh crore. This basically means that during the period under consideration 124 per cent of the money that came in as bank deposits was invested into government securities. Hence, not only all the deposits that came in between October 28, 2016, and January 20, 2017, have been invested by banks into government securities, some of the earlier deposits have also been invested into government securities.

Given that cash is fungible, this basically means that on the whole banks haven’t loaned out any of the deposits that have come after demonetisation. All that money has been invested into government securities.

Now let’s take a look at Figure 2. This essentially plots the portion of bank deposits which have been given out as loans (i.e. non-food credit). Banks give working-capital loans to the Food Corporation of India to carry out its operations, those have been adjusted for.

Figure 2: 

What does Figure 2 tell us? It tells us that before demonetisation banks had loaned out 73.3 per cent of the deposits. This has since fallen to 69.7 per cent. This isn’t surprising given that banks cannot immediately lend out all the money that has come in.

There is another question that needs to be answered here. Given that interest rates have fallen because of demonetisation, how much has bank lending gone up by, before and after demonetisation? Between October 28, 2016, and January 20, 2017, lending by banks went up by a minuscule 0.45 per cent. This basically means that the lending has more or less been flat.

Also, given that all the deposits that have come in since October 28, 2016, have been invested in government securities, this essentially means that some of the deposits that had come in earlier, have been lent out.

This, also tells us, all over again, that lower interest rates are not the only factor that leads to increased borrowing. Hence, the theory of lower interest rates leading to increased borrowing leading to better economic well-being, due to monetisation, does not really work. The data does not show that at all.

There is another point that needs to be made here. A significant amount of investments made by banks in government securities has been made under the market stabilisation scheme(MSS). The government securities issued under the market stabilisation scheme has all the characteristics of regular government securities. But, unlike the regular government securities, these securities are not issued to finance the fiscal deficit of the government. Fiscal deficit is the difference between what a government earns and what it spends and is financed by issuing financial securities referred to as government securities.

What this basically means is that money borrowed under the market stabilisation scheme lies idle. It isn’t used to finance the expenditure of the government. On December 2, 2016, the RBI increased the ceiling of the government securities that could be issued under the market stabilisation scheme to Rs 6 lakh crore. Earlier the limit was Rs 30,000 crore.

As the RBI pointed out in a press release: “After the withdrawal of the legal tender character of the Rs 500 and Rs 1000 denomination notes with effect from November 9, 2016, there has been a surge in the deposits with the banks. Consequently, there has been a significant increase of liquidity in the banking system which is expected to continue for some time.”

The government securities issued under the market stabilisation scheme sucked out this liquidity. But this money has been lying idle with the RBI. Earlier it was a part of the financial system and was helping people carry out transactions. This has reduced the velocity of money. And a lower velocity of money leads to lower economic activity.

Over and above this, the surfeit of deposits coming in has led to banks slashing the interest rates on their fixed deposits. Take a look at Figure 3.

Figure 3: Repo, Base Lending Rate and Term Deposit Rate (Per cent) 

Let’s analyse Figure 3 in some detail. The base rate is essentially the interest rate below which a bank cannot lend i.e. the interest rate at which a bank lends to its best customer. The term deposit rate is essentially the interest rate that a bank pays on its fixed deposits.

Since January 2014, the term deposit rate of banks has fallen. At the same time, the base rate has also fallen. Nevertheless, the term deposit rates have fallen much more and at a far greater speed than the base rates.

This is something that becomes clear by looking at Figure 3 carefully. The gap between the average base rate and the average term deposit rates has increased considerably between January 2014 and December 2016. The base rate has barely moved from 10 per cent to around 9.5 per cent. On the other hand, the term deposit rate has moved from around 8.5 per cent and is now below 7 per cent.

As the Economic Survey points out: “By December 2016 the gap between the average term deposit rate and the average base rate had grown to 2.7 percentage points, from 1.6 percentage points in January 2015.”

The question is why have the banks done this? Over the longer term, the banks have been trying to make up for their bad loans by doing this. As the Economic Survey points out: “They have tried to compensate for the lack of earnings from the non-performing part of their portfolio by widening their interest margins.”

Over the short term, they are simply doing this because of the surfeit of deposits that have come in. They have invested this money in government securities (a large part), which do not pay a high rate of return like lending that money out would. To compensate for this, the banks have cut interest rates on their fixed deposits faster than the interest rates on their loans.

Hence, notebandi or demonetisation has led to lower interest rates on fixed deposits. A major part of the household financial savings in India are held in the form of bank fixed deposits. Anyone looking to meet an investment goal now must save a greater amount and this will leave a lower amount of money for consumption.

Further, lower interest rates haven’t led to a higher lending by banks. All this has led to is money lying idle with the RBI. And that is something that India cannot afford.

(The column originally appeared on Equitymaster on February 16, 2017)

What You See is All There is?

daniel_kahneman

One story that constantly get splashed across Mumbai tabloids (the city I live in) is that of chain snatching. Almost all such newsreports give a feeling that the city is no longer as safe as it used to be.

But is this true? As Daniel Levitin writes in A Field Guide to Lies and Statistics: “We assume newspaper space given to crime reporting is a measure of crime rate. Or that the amount of newspaper coverage given over to different causes of death correlates to risk. But assumptions like this are unwise.”

Psychologist Daniel Kahneman calls this phenomenon “what you see is all there is” or WYSIATI. As he writes in Thinking, Fast and Slow: “The confidence that individuals have in their beliefs depends mostly on the quality of the story they can tell about what they see, even if they see little. We often fail to allow for the possibility that evidence that should be critical to our judgement is missing—what we see is all there is.”

Hence, what we believe in tends to get decided by what we see, and in the process, we tend to ignore the evidence that we should be really looking for. Take the case of the chain snatching in Mumbai that I talk at the beginning of the column. Any standalone chain snatching incident on its own should not lead us to draw broader conclusions. Nevertheless, that is precisely what we do, concluding that the city is more unsafe than it was in the past.

The data that we should basically be looking at is: Whether the number of chain snatching incidents has gone up than in comparison to the past? And if that is the case, then it is fair to draw the conclusion that the city is more unsafe than in comparison to the past, at least on the chain snatching front.

But this is a data point that most media reports on chain snatching tend to miss out on. A recent newsreport points out that the number of chain snatching incidents reported to the police in Mumbai, has fallen from 1,954 in 2012 to 445 in 2016. Also, the proportion of cases where a detection has been made has increased from 36.7 per cent in 2012 to 55.1 per cent in 2016. This means that a greater portion of the chain snatching crimes have been solved.

As per these figures, Mumbai is safer on the chain snatching front than it was in the past. But given that these data points rarely get reported that is not the impression that people have of the city. As Levitin writes: “Cognitive psychologist Paul Slovic showed that people dramatically overweight the relative risks of things that receive media attention. And part of the calculus for whether something receives media attention is whether or not it makes a good story. A death by drowning is more dramatic, more sudden, and perhaps more preventable than death by stomach cancer—all elements that make for a good, though tragic, tale.”

Given this, drowning deaths tend to get reported more than deaths by stomach cancer. This  leads people to erroneously believe that they are more common, which is not the case. As Levitin writes: “About five times more people die each year of stomach cancer than of unintentional drowning… Misunderstandings of risk can lead us to ignore or discount evidence we could use to protect ourselves.”

And that brings us back to the original question, is what you see really in the media all there is? Now that is something worth thinking about.

The column originally appeared in the Bangalore Mirror on February 15, 2017

 

 

India’s Ecommerce Ponzi Scheme Has Started to Unravel

flipkartA spate of newsreports in the recent past clearly show that Indian ecommerce companies are in trouble.

A newsreport on Moneycontrol.com points out: “With an aim to cut costs, struggling e-commerce firm Snapdeal is likely to downsize its team by around 1,300 employee.” This is around one-third of the company’s total workforce of 4,000 employees.

On the other hand, Flipkart has shutdown its courier service and hyperlocal delivery project, less than a year after launching it. There are other examples as well. The question is why are companies doing this? They are trying to cut down their costs and at the same time conserve on all the money they have raised from investors.

Over and above this, investors have made a spate of mark-downs to their investments in these firms. A January 27, 2017, newsreport on Reuters points out that Fidelity Investments has marked down its investment in Flipkart by around 36 per cent. In December 2016, Morgan Stanley, had marked down its investment in Flipkart by 38 per cent.

The Japanese investor Softbank recently marked down the combined value of its shareholding in Ola and Snapdeal by $475 million. What does all this mean? It essentially means that these investors do not accept these ecommerce firms to be as successful as they expected them to be in the past. And given this, they have been writing down the value of their investments.

In a column, I had written early last year I had called Indian ecommerce firms a Ponzi scheme. Of course, this had led to a lot of abuse on the social media and I was told that I do not understand the business model of these firms. I wrote what I did because I understood the business model of these firms. Allow me to explain.

A look at the profit and loss numbers of these firms will tell you that the losses of these firms go up at the same time as their revenue.  Take the case of the market major Flipkart. As a report in the Business Standard points out, for the financial year ending March 31, 2016, the losses of the firm stood at Rs 2,306 crore. The company’s losses for the year ending March 31, 2015, had stood at Rs 1,096 crore. Where did the revenue of the firm stand at? It jumped from Rs 772.5 crore to Rs 1,952 crore, during the same period.

Or take the case of Snapdeal run by Jasper Infotech Private Ltd. A report in the Mint points out that for the financial year ending March 31, 2016, the losses of the firm stood at Rs 3,316 crore. For the financial year ending March 31, 2015, the losses had stood at Rs 1,328 crore. During the same period, the revenue of the firm increased from Rs 933 crore to Rs 1,457 crore.

What sort of a business model is this—where the losses of a company go up at the same time as its revenue? In fact, in case of Snapdeal, the losses have gone up at a much faster rate than its revenue.

What explains this basic disconnect? As Gary Smith writes in Standard Deviations—Flawed Assumptions, Tortured Data and Other Ways to Lie With Statistics: “A dotcom company proved it was a player not by making money, but by spending money, preferably other people’s money… One rationale was to be the first-mover by getting big fast… The idea was that once people believe that your web site is the place to go to buy something, sell something, or learn something, you have a monopoly that can crush competition and reap profits.”

What was true about American dotcoms is also true about Indian ecommerce companies. This isn’t surprising given that many investors in Indian ecommerce firms are American.

I discovered Flipkart one day in 2009. Back then it was simply an online bookstore. It had a reasonably good collection of books. It even had books which bookstores did not. And the deliveries were on time.

What else did one want? Discounts. It had good discounts on offer as well. Hence, out went the bookstore and in came Flipkart. The loyalty was to discounts and nothing more. Sometime later, when other websites like Homeshop18 and even Amazon, started offering higher discounts, I moved to ordering from these websites.

Nevertheless, one did wonder, how would these websites ever get around to making money, given the huge discounts that they offered. The way businesses run traditionally it never makes any sense to sell a product below the cost all the time, because that way the business is never going to make any money.

But these websites did not fit into the traditional way of doing things. At least, that is way they thought. The best way to explain this is through the example of a telephone. As James Evans and Richard L. Schmalensee write in Matchmakers: The New Economics of Multisided Platforms: “A telephone was useless if nobody else had one. Even Bell and Watson started with two. A telephone was more valuable if a user could reach more people.”

The point being that more the number of people who had a telephone, more the number of people who would want to have a telephone. The economists call this the phenomenon of the direct network effect. This essentially means that more the number of people who are connected to any particular network, the more valuable it is to people who are already a part of it.

Take the case of app-based cab services. When they launched, they offered rock bottom rates. This was done to attract customers. Once customers came on board, it was easy to attract more and more drivers on to the network as well. And over a period of time, the price of these app based services has gone up.

Of course, it is not easy as I make it sound. But that is the basic logic. Then there are apps which deliver food from restaurants. They also offered discounts initially in order to build a critical mass of customers to be able to attract good restaurants on the platform.

As economist John Kay writes in Everlasting Last Bulbs—How Economics Illuminates the World: “The company that is first to create the largest network denies access to competitors and establishes an unassailable monopoly…Connectedness is vital, and it is best to be connected to the largest network.”

So, the ecommerce game is centred around building a monopoly and cashing in on it. As Ray Fisman and Tim Sullivan write in The Inner Lives of Markets in the context of network externality: “The bigger a company gets, the more valuable it is to each successive customer, there’s a huge premium on expanding your customer base.”

And this explains the discount led model. In case of Flipkart, the discount led model was first offered on books to build a critical mass of customers, and then the company gradually got into selling many other products. The hope was that once the consumer was comfortable buying books from the website, he would become comfortable buying other products as well, over a period of time. The logic worked on the supply side as well, as more and more vendors got comfortable selling online, more vendors came in.

Of course, the discount led model leads to losses. Hence, any company following this model, needs money from investors to keep running. And this is where the structure of Indian ecommerce companies becomes similar to that of a Ponzi scheme.

A Ponzi scheme is essentially a financial fraud in which investment is solicited by offering very high returns. The investment of the first lot of investors is redeemed by using the money brought in by the second lot. The investment of the second lot of investors is redeemed by using the money brought in by the third lot and so on.

The scheme continues up until the money being brought in by the new investors is greater than the money being redeemed to the old investors. The moment the money that needs to be redeemed becomes greater than the fresh money coming in, the scheme collapses.  How does this apply in case of Indian e-commerce companies?

Indian ecommerce companies have managed to survive because of investors bringing in fresh money into the scheme at regular intervals. It is worth mentioning here that every time investors bring in more money, they bring it in at a higher valuation. This essentially means that the price at which shares of the company are sold to the investors are higher than they were the last time around. This increases the market capitalization of the company.

This increase in market capitalization comes about because the company has managed to increase its revenue. As long as the money being brought in by the investors keeps subsidising the losses being accumulated by the e-commerce firms, these firms will keep running. The moment this changes, the firms will start to shut-down. The structure of the Indian e-commerce companies is that of a classic Ponzi scheme.

Nevertheless, as we have seen earlier in this column, this increase in revenue typically comes at the losses increasing as well. This is a fact that investors of these firms have started to realise as well. And that is why they have marked down the value of their investments.

An investor who is marking down his investment is unlikely to invest more money into the firm. If he actually goes about investing more money in the firm, then he is likely to do it at a much lower valuation. Given this, the Indian Ecommerce Ponzi scheme is now unravelling. The trouble is that everyone wants to be build a monopoly. But everyone cannot be a monopoly.

As Smith writes in the context of the American dotcom bubble: “The problem is that, even if it is possible to monopolize something, there were thousands of dotcom companies and there isn’t room for thousands of monopolies. Of the thousands of companies trying to get big fast, very few can ever be monopolies.”

This basic logic applies to the Indian ecommerce as well. And given this, if fresh investor  money stops coming into these firms, as it has in many cases, these companies will soon start going bust.

To conclude, it’s time we got ready for the ecommerce bloodbath.

The column was originally published on Equitymaster on February 15, 2017

And the Notebandi Lies Continue…

Fostering Public Leadership - World Economic Forum - India Economic Summit 2010

One week back, the finance minister Arun Jaitley said in the Rajya Sabha: “At no point of time, not for a single day, was the currency inadequate.”

Every government needs to defend decisions it has taken. In that context Jaitley’s statement is hardly surprising. Nevertheless, it not only mocks the common man of this country but also tells us how disconnected our ruling politicians are with the realities of the day.

Jaitley was essentially talking about the situation that prevailed in the aftermath of demonetisation or notebandi, as it is more commonly referred to as.

If there was no currency shortage for even a single day, why did ATMs have such long lines for close to a month? Is Mr Jaitley saying that people just gathered there because they had nothing else to do?

If there was no currency shortage even for a day, why did the government place limits on ATM withdrawals? It was churlish of Jaitley to have said what he did, given that 86 per cent of the currency in circulation was demonetised, the midnight of November 8, 2016, onwards.

The advantage with making speeches is that nobody asks questions at the end of it. Nevertheless, the lack of empathy among the politicians does get registered.

That apart, let’s look at the currency in circulation data published by the Reserve Bank of India every week. Take a look at Figure 1.

 

Figure 1:

Figure 1 essentially shows the currency in circulation in the Indian economy in 2017. December 30, 2016, was the last date for depositing the Rs 500 and Rs 1,000 notes which had been demonetised. Hence, I have taken the currency in circulation numbers from January 6, 2017, onwards, which is a week later.

The currency under circulation has been going up since 2017. On November 4, 2016, four days before prime minister Narendra Modi, made the announcement to demonetise Rs 500 and Rs 1,000 notes, the total currency in circulation had stood at Rs 17.97 lakh crore. In comparison, on February 3, 2017, the total currency in circulation, the latest data available, stood at Rs 10.49 lakh crore.

Hence, the total currency in circulation as on February 3, 2017, was at 58.4 per cent of the level before monetisation was announced. Given this, it is not surprising that the currency shortage, even though it has eased, continues to persist. This goes against what the Economic Affairs Secretary Shaktikanta Das recently said about the remonetisation process being complete.

Take a look at Figure 2. It basically plots the total increase in currency in circulation every week since January 6, 2017.

Figure 2:For the week ending January 13, 2017, the total currency in circulation grew by Rs 52,780 crore. Thereafter, the increase in circulation has fallen quite dramatically. One explanation for this may lie in the fact that initially more Rs 2,000 notes were being printed and that has now been replaced with more Rs 500 notes being printed, which is what the financial system needs, given the shortage of change. But at the same time, it takes four Rs 500 notes to replace money worth Rs 2,000.

The larger point being that the financial system is still away from having an adequate amount of currency. This, as I have explained in the past, is primarily because of the limited currency printing capacity of the government of India and the Reserve Bank of India.

The average increase in currency between January 6 and February 3, 2017, comes to around Rs 37,778 crore. At this speed, it will take many more weeks, before the financial system gets to a level, where it has adequate currency.

The total currency in circulation had stood at Rs 17.97 lakh crore before demonetisation. As of February 3, 2017, the total currency in circulation stood at Rs 10.49 lakh crore. The difference between this and the total amount of currency in circulation before demonetisation stands at Rs 7.48 lakh crore (Rs 17.97 lakh crore minus Rs 10.49 lakh crore).

At the speed of introducing currency worth Rs 37,778 crore per week, it will take close to 20 weeks for the currency under circulation to reach the pre-demonetisation level. One logic that has been offered is that the government may choose not to replace the entire currency.

Even if the government chooses not to replace the entire currency, at Rs 37,778 crore per week, it will take many more weeks before the currency in circulation stabilises at an adequate level.

While, the economics of it, can get tricky, even if the government chooses to go up to Rs 16 lakh crore and not Rs 17.97 lakh crore, it will still take close to 15 weeks to get to the pre-demonetisation level. Of course, the time taken can come down if the speed of money printing can be increased.

Long story short—both Jaitley and Das are essentially lying to the country in saying what they are. And that is something worth remembering and talking about, dear reader.

The column was originally published on February 14, 2017, on Equitymaster