Should Google be worried by a conservative lawsuit

Megan McArdle writes a well thought out post here about there being no upside in being sued by an ex-employee who claims conservative views are grounds for discrimination. I usually agree, but would like to play devil’s advocate on this one.

Her argument boils down to two points and one piece of supporting evidence.

  1. Google exposes itself to boycotts from businesses which will affect its revenues and profits
  2. Voters and politicians who lean conservative will punish the bias (even if its only perceived and not real) and push for regulating the previously “not evil” Google.

The supporting evidence — most mature companies do not publicly lean towards a particular political side.

My argument is quite simple.

  1. While the likelihood of punishment from regulators is not negligible, its unlikely to be purely aimed at Google alone. Instead it will be aimed at the industry. And in most industries with reams of paperwork to wade through, it is the incumbent that wins.
  2. Google wins with a better product which is a stranglehold it is yet to give up. This is primarily due to its employees. So far, its method of hiring appears to have worked. We need evidence that it stops being an employer of choice before we start ringing death knells.

Having said this, I am not comfortable with the idea of firing an employee, however odious his views may be. But I am also not certain working with an employee who is publicly open with his views that half the human race is incapable of doing his job at his level of skill is an option either.

With respect to supporting evidence, I tend to be very wary of the view that mature companies do not choose sides. Perhaps it is my cynicism seeping through, but I think its more because they are better at hiding their political leanings than anything else.

Management, Telecom

Net Neutrality — A free market view

There has been a lot of kerfuffle and reams of digital newsprint wasted over this strange phrase “net neutrality”.

Some people seem to say that without it, we would not have the internet anymore and all that remains would be cat memes and twitter trolls. Others seem to say that it is the one thing that stops the world from giving internet to the starving masses huddled in rural corners of the world. Somewhere in between these two views, the truth seems to have managed a vanishing act.

Now I approach most economic choices from viewpoint that people should be allowed to do whatever they want (As long as they don’t swing their arms right onto my nose). So let us try to imagine a few scenarios.

Warning: This is a long post and needs to cover a whole series of ideas. Here be many dragons, and most of them this writer has yet to scratch, let alone slay.

Part I — Telecom industry

Scenario number 1: Anything goes.

In this scenario, everything is a free for all. Everybody can open a telecom company with nothing more than a dream and sales pitch. Would it work? Not really. Why? Because of this unfortunate constraint of limited wireless spectrum for the wireless guys. And because laying wired internet through underground cabling is EXPENSIVE. You need more than a dream and a sales pitch.

When anything goes, people will make a beeline for scarce spectrum and squat on it, because it is valuable. Assuming that this anarchist utopia somehow also respects property rights, those with the spectrum rights would hoard this and charge the the maximum they can get away with. However, because there would ideally be a lot of small players grabbing spectrum, the amount of pricing power each of the small players would have would be quite small.

What would the downside be?

  • Calls between two different networks would be hit or miss. With several different networks all competing with each other, it is very possible that some would not be up to scratch
  • Profitability of operators: The joy of free markets is greatest when there are a very large number of service providers. However, the downside is that you get spectacular price wars. This is great for customers, but terrible for investors who will forever be on the edge of profitability.
  • Investment: Which neatly brings up to the third point. Wireless telecom is still very fast changing. Speeds are available today which were not even dreamed of even a decade ago. Unfortunately, all this costs money. Firms which are marginally profitable would have great difficulty in investing for the long term.

Caveats: All of these downsides are solvable by enough free markets and time.

This is basically the scenario that faced telecom service providers in India in the mid 2000s (not really, but for our example it will serve). From a state monopoly in 1992 to 15 mobile phone operators in India by 2008, the industry had come a long way. But not the operators. Barring Airtel, which had been in business from the beginning most of the others were engaging in a brutal race to the bottom, with tariffs crashing from INR 24 per minute to INR 0.5 per minute. At the same time, technology had to be upgraded from 2G to 3G then HSPA and 4G.

We will pick up this story later in this article.

Scenario 2: Monopoly

The opposite of free markets is state control. I was barely into my teens when telecom monopolies were gradually pulled down, but even I remember the long wait for a telephone connection from the only provider there was. The downsides of monopoly, especially state controlled monopoly is one that is very easy to identify.

  • The lack of the profit incentive means there are no objective ways to measure performance. This makes rational employees prioritize managing official relationships over managing customer relationships. This leads to customers getting terrifyingly bad service
  • Lack of competitive alternatives leads to stagnation in development of new technologies or marketing strategies. Why change if there is no upside?
  • No profits: Traditional monopolies tend to throw out cash. But Public sector monopolies cannot afford to be profitable because that would indicate price gouging, which would lead to an unpopular government. And no government would like to be unpopular.
  • High operating expenses: Paradoxically, the lack of profits does not mean services are cheap. Instead, the surplus that a monopoly extracts from the customer goes to the third stakeholder: the management. Management builds empires, goes for moonshots, build legacies.

I am pretty sure I do not want to go back to a world where BSNL is the only service provider. Scenario 2 was the structure of Indian telecom since at least the early 80’s. One hopes those days never come back.

Scenario 3: The regulated free market with spectrum auctions + revenue sharing (after some years.)

Now this is approximately where we got to in the later part of the last decade. When 3G spectrum was to be released to the Indian telecom companies, the government decided to try something that was done at the end of the millennium in parts of the developed world: The Auction (Aside: This also marks my first attempts to follow business news.) For an excellent report of that event see here The result was a bonanza for the government. Over $20 billion dollars of revenue came in (mostly) one fell swoop. This also had the unfortunate side effect of making people sit up and wonder how much the state gave up when it gave away 2G spectrum along with telecom licences for a pittance.

The hypercompetitive players from scenario 1 had time before 3G became a thing. It was only by 2010 that the reality of the situation began to dawn on the players. The first to capitulate were the marginal players like Telenor and MTS. But the final blow was still to be struck.

Enter the 800 pound gorilla

India’s largest diversified conglomerate is Reliance Industries Limited (RIL). It has a huge cash cow in its oil refinery business that throws of cash by the billion every quarter. By 2012, it was clear that RIL was out to re-enter the telecom business (it had a telecom business, Reliance Communication, which had been given to the other brother in a family dispute). By opting for a fully 4G network and offering Voice + Data in an integrated manner, it had the capability to truly alter the digital landscape. But for 4 years, there were only rumours. RIL always said they were building up infrastructure and it was not ready to go to market yet.

This all changed in late 2016 and early 2017. RIL entered the game with a bang. With data being the only thing people had to pay for, and voice being free it squeezed the smaller players like Reliance Communication, Aircel, and Tata Docomo. These were already barely scraping by and this was the final nail in the coffin. Reliance and Aircel went on to explore a merger which appears to have imploded. The Tatas have sold their operations to Airtel, while Idea and Vodafone are merging to bulk up. Eventually it appears that India will finally have what the developed markets have had. 3 major players and a couple of minnows.

All this history serves to explain something which the knowledgeable reader already knows. Companies need to remain profitable to provide services. It may seem obvious, but it appears that many on either side of the net neutrality divide appear to forget this.

Part 2 — Net Neutrality

Now we come to the topic of the moment. First, what is net neutrality? One sentence definition: All content in the internet should be treated equally. In theory this means that my hypothetical bit-torrent download of the latest episode of Game of Thrones (No. I do not watch it. I still forlornly wait for the books) should load just as quickly as Ms. N’s cat videos on Facebook. Is this currently the case? The telecom service providers insist that it is. Based on my anecdotal experience, it is almost certainly not true.

But if this were all the fight was about, it would be fought by some tech guys in hoodies along with some engineers in telecom companies. But its not only about downloads. There are a few things to conside

1. Netflix and Video Downloads

When someone opens Netflix and starts to watch the latest season of “The Crown”, how does it get to your screen? It starts its journey on a Netflix server somewhere in the world. Since Netflix video content is so bandwidth intensive, if enough people in your apartment are watching it simultaneously your local network will start feeling the strain. Streams will start to buffer if your ISP did not expect everyone in your building to stream 4k video at the same time. If it happens often enough, you will call your ISP and tell him/her that you are going to switch providers. ISP’s don’t like that too much so they will then be forced to install more infrastructure to feed your Netflix habit. As you may imagine, they feel a bit aggrieved and would like Netflix to defray some of that cost. As things stand today, they should not be allowed to ask this. Why? Because carriers are supposed to be content agnostic. It should not matter to them if you are streaming netflix video or video from youtube, or even if you are streaming the backup copy of your steam library (currently at over 200Gb).

We will look at this argument first, as its quite easy to put down. Netflix should not be made to pay twice. They are paying their ISP anyway. Getting the links between their ISP to the customers computer is what you, the consumer are paying your ISP for. If your provider (say Airtel Broadband) is not able to convince you to pony up more money for the privilege of high speed UHD video, then surely Netflix should not be made to subsidize your yen for it. If Airtel Broadband can make those investments profitably with you as the paying customer, then so be it. Else, they can abandon you to your fate of buffering video (and your switch to ACT Fibernet if in Bengaluru).

2. Facebook Zero and the walled garden

This is a case which is more complicated. A hypothetical Facebook zero encapsulates it, where you get free Facebook and Whatsapp which will be sponsored by Facebook. But because the consumer does not want (or can afford) anything more, he/she will not be allowed access to anything else (like Wikipedia). Is this truly such a bad thing? Could we not allow for independent businesses to slice and dice the internet in this way? The wisdom of the marketplace would suggest that this should be tried out, and if the internet has truly to be transformed into Facebook likes of cat videos, then so be it. If people believe that this is not a valuable tool, then perhaps a LinkedIN zero will be more remunerative.

My own take on this is slightly more nuanced. I see no real problem with a walled garden commercially, though my personal views on “Free Amazon” is that anything that is too good to be true probably is. However, this sort of slicing and dicing also imposes costs on consumers. These costs are borne in variety of plans that ISP’s would necessarily need to increase. One example would be the “Facebook + LinkedIN zero plan” or perhaps a “Youtube + Netflix” one. In the quest for customization, the original internet plan of 500 rupees for 30 Gb of data may well fall by the wayside, because the market for Steam download plan or Linux distribution download plans are just too niche for the market to ever provide such plans. Personally, I think this would be a low probability outcome, but one that is still possible.

Should regulators ban such commercial transactions? I honestly think that an Amazon only product is anti-competitive enough that a Flipkart should feel aggrieved (and vice versa). I also believe that such products would mean the end of new startups as the barriers to entry would be enormous in the form of capital requirements. While these sort of plans may help the telecom market through higher revenues from marginal customers, the long term impact on the broader economy is definitely negative. I also think that the best way to learn is through mistakes, and setting up such plans will demonstrate their folly.

3. Fast and slow lanes for services

This one is where I tend to explicitly disagree with the net neutrality theory. Some services require very low latency or very high bandwidth. An example would be self driving cars, or robotic surgery. You need to have the capability to prioritize such bandwidth which is “Socially Useful and Productive”. How does one decide what is socially useful, and what is fluff? Again, I am not sure who will, but I am totally sure the answer is not “Free markets”. This is an area where thoughtful regulation can actually provide gains to both industry as well as the broader society that exists around it.

There are other issues which while being peripheral to the main net neutrality case are not insignificant. One of them is deep packet inspection and the role of ISP’s in monitoring consumer consumption of data. I will leave it to the reader to learn more about such things (not to mention, my own knowledge of this is sorely lacking).

We live in a world of blacks and whites. The idea that regulation is a spectrum and not a bright line is one that has lost believers.

Part 3 — Regulators, Revenues and Auctions — Is it a free market at all?

Now, the careful reader has obviously noted something. Part 1 and Part 2 are totally different topics. They have had nothing to do with each other. Why bundle them up into the same post?

Because the whole net neutrality debate is also one that affects telecom company bottomlines. Companies that are on the brink of solvency are justifiably concerned that the chaos of new business models may tip them over. Companies that are already over the brink (Looking at you, RCOM) have nothing to lose. While a company like RJio, that is capable of massive investments in content as well as business models is probably salivating at the prospect of bundling services in a fully walled garden.

India’s net neutrality guidelines have very neatly kept the door of intranet content bundling ajar while batting for net neutrality across the remainder of the internet. I suspect that that little opening will be enough for a full scale content war. The outcome of this war is uncertain. RIL and its free cash flow can fund an enormous amount of content, while Airtel and Vodafone are also pushing forward. The future will however be decided by the consumer, who is now used to not paying a high price for any data.


General stuff, Politics, Uncategorized

Crackers and the judiciary

I rarely write on judicial orders, but N is tired of me ranting to her, so am indulging myself on the web.

The supreme court in its wisdom has decided to ban the sale of firecrackers in Delhi due to the (quite justifiable) fear that the pollution caused will result in health issues. Given Delhi’s quite abominable air quality, the fact that adding a bunch of Sulphur di-oxide is going to be quite deleterious is indisputable.

What is disputable however, is the fact that while sale is now illegal, it is less certain that actually bursting fire crackers is illegal. I am almost certain that it is not illegal in most of the rest of the country. And by making sale of firecrackers (otherwise legal in most of India) is the court not infringing on the rights of all those otherwise upstanding businessman to conduct their trade?

Now, I have no skin in this game. I neither live in Delhi, nor am involved in the cracker supply chain in any form or fashion; being an LED diya person. But the trend of judicial activism into areas which traditionally belonged to government and legislative action is something that is increasingly visible in India. It may have had benign intentions and may have been widely lauded, as when the court ruled that buses in Delhi should only run on CNG and not diesel. But this has emboldened the courts to bravely venture forth where no judge has gone before.

From crackers to alcohol shops, from coal licences to telecom spectrum, the reach of the legislature has shrunk, with the judicial arm of the state taking its place. We may not like our politicians, and we may disagree strongly with their actions. But I think its worth asking if the courts taking their place is a good solution.

Banking, Economics, Uncategorized

Public Sector Banks – status report

Over the last few months, the Indian banking system has shown its true colours. And unlike the Phil Collins song, they aren’t beautiful at all. With gross bad loan numbers ranging from 7% (Indian Bank) to 24% (IDBI Bank), the best of them need significant capital infusion to grow, and the worst are on the verge of insolvency.

To paint over the true colours, the government had announced Indradhanush (Hindi for Rainbow), a recapitalization of banks which would total INR 700 trillion. This amount was frontloaded with INR 300 Trillion being infused into the banks in 2015-16. The remaining 250 trillion was disbursed in 2016-17. Now there is only 100 trillion every year to provide for the next 2 years. Sadly for the banks, this  is not even remotely going to be sufficient. Let us examine why?


Non Performing Asset – Types and Introduction


There are two types of  Non Performing Assets (NPA). The first, called gross NPA is made up of all the loans of a bank where the interest is not paid for over 90 days. The second, called net NPA, reduces the gross NPA level by the amount of money the bank has set aside to cover for these non-paying loans. This money is called provisions. There are also some minor adjustments, which we will leave out for purpose of simplicity. So why do we have 2 types of NPA’s? Why can’t one do?

Gross NPA

The gross NPA number is a good indicator of current earnings of the bank. A banks business depends on people paying interest. So, the percentage of loans that don’t pay interest is useful when people want to look at future earnings.


The net NPA number is a good indicator of the solvency of the bank. Banks are quite odd creatures. They take in deposits and lend those deposits out to other people as loans. Banks must pay back those deposits, even if the loans aren’t paid back. They can do this either from earlier years profits, or from the capital that the bank has taken from its shareholders.


The NPA problem for PSU Banks

Now that we know some of the jargon that bankers use, let us look at some numbers. The total Gross NPA of the PSU banks stands at 7.29 trillion rupees. An estimate of their total loans is approximately 57.17 trillion rupees. So effectively one eighth of the loans a bank gives out is no longer paying any interest.

The net NPA number is a little more complicated as there are some adjustments made by each bank, but the aggregate number is approximately 4.42 trillion rupees. This number needs to be adjusted against the shareholder capital and any reserves and surplus that the banks may have. The total shareholder funds of all the PSU banks is about 5.817 trillion rupees. This means that if all the PSU banks were considered as one megabank, they would have additional shareholder capital of about 1.4 trillion rupees.

Do note that in this discussion, there is no mention of the regulatory capital requirements that are mandatorily needed. The current total loans is approximately 50 trillion Rupees (after subtracting the bad loans). The percentage of equity remaining to total loans is approximately 2%. The regulatory requirement I have seen for Tier 1 capital is 5%. Effectively, the PSU banks in aggregate have to raise at least 2.5 times their existing shareholder funds in order to meet the minimum requirement to be called solvent.


A large number of trees have been killed and whole reams of newsprint been used to discuss the NPA problem in PSU Banks. This article adds to that discussion, without proposing a solution. This is deliberate. I believe that defining the problem accurately is the first step to actually solving it. In further articles, we will take a look at the possible alternatives available to the regulator and the principal shareholder, the Indian government.


Economics Post — Information asymmetry in free markets

Over the last several months I have been thinking about several factors that seem to affect the behavior of prices of goods and financial instruments, and realised that they all seem to revolve around the concept of information asymmetry.

There are a lot of points to be covered, so this may be part of a multi post series.

Information Asymmetry — What is it and why is it important?

In any marketplace, the following conditions need to be met for a perfect market.

  1. No individual market participant can control pricing (practically infinite market participants)
  2. No barriers to entry or exit (No significant capital costs or bankruptcy coss)
  3. Non-intervention by external forces (eg: No government intervention or weird taxes on products/services)
  4. Profit maximisation (parties behave rationally and the measure of rationality is profit)
  5. Perfect market information (every market participation has all information available about the product/service)

A liquid stock market satisfies requirement 1 and 2. Item 3 is a bit more problematic, but we can assume it to be irrelevant for the purpose of this discussion. 4 is definitely a problem area as human beings are rarely rational, but will have to be tackled at a different point in time.

Perfect market information is an interesting one though. This is one that should be easy to establish. If there is any information asymmetry about a stock, those who possess the information should be able to exploit this asymmetry and thereby move the price. Hence the phrase, ‘the news is in the price.’

However, this is not true. In fact if it were perfectly true, then everyone would agree on the price of a stock. And once they agree on the price, what is the incentive to buy or sell it? Warren Buffet thinks that Wells Fargo is a good company that motivates its employees to cross sell products, so he would like to buy it. And Elizabeth Warren thinks it is a fraudulent company that is built on exploiting poor people and forcing them to cross sell useless products to customers. So she would gladly dump the stock (assuming she owned it). The news in this case is identical. But the interpretation is very different.

There are other instances which are less simple. Lets take the example of a retail investor who sees that Walmart is up 10% for the year. He is pretty happy with this gain and wants to sell his shares. A hedge fund owner on the other side has invested 10 million dollars to buy time on a satellite camera to look at all the parking lots at all the Walmarts in the United States on Thanksgiving weekend. He sees that the car parks are all full. His camera even spots that the boots of the cars hold huge flatscreen TV’s. So the hedge fund guy (and its almost always a guy) will be happy to buy Walmart stock from the retail investor because he knows that Walmart quarterly numbers next month will look great.

Is using a satellite to get an edge unfair? Sure it is! But nobody said the market had to be fair. It only has to work.

To summarise, information asymmetry is a key part of what makes free markets work. However, the very thing that makes the market work also makes it deeply weighed in favour of those with the resources to get the edge that makes them profitable.

Future posts will take this thought forward a bit more.

Economics, Freakonomics, free trade

Productivity measured through cost alone

Productivity improvements are almost always considered to be an unmitigated good. They help the overall economy grow by making more output out of a finite input. As an example, the move to computers from typewriters was quite a good productivity improvement. (Most) People could type documents faster with a computer than they could with a typewriter.

However, it is interesting to see how productivity is measured and managed today in the globalized single market. As an example, let us take software development. In the world pre-globalization and outsourcing, the cost to develop and maintain software in the US was a function of the price you had to pay a programmer in the US.

Today, you can hire a programmer from India to do the same thing at a third of the cost. And simple exchange rate and cost of living arbitrage means that the Indian programmer would be able to have a better standard of living even at this spectacular lower cost.

This seems like a no-brainer to most companies today, and lots of jobs are being shunted around to low cost locations around the world leading to improved margins or dramatically lower prices. However, this spectacular improvement in productivity for the global economy as a whole has some very pertinent local impacts.

Developed markets which have relatively high labour costs tend to have their jobs threatened by these low cost interlopers. In the 1980’s and 90’s this was the skilled manufacturing jobs that moved to China and South East Asia. In the 2000’s and this decade, the shift has been less labour intensive, but no less profound as low value IT jobs and back office processing moved to India and the Philippines.

The next steps will be a shift when relatively high value jobs, like management consulting and legal and financial advice get shifted to the lower cost countries.

A careful observer will state that these shifts are actually a good thing. Today, Apple is the most valuable company in the world. And while its products may be made in China, they are designed in the United States and the maximum value (30% margins) that is generated is reaped within the United States, with China getting the crumbs in the form of manufacturing margins of 3-4%.

However, this means that the developed world labour markets are in constant flux, with the employees in this world needing to constantly upgrade their skills or to perish in the low margin world of the outsourced.

Is this sustainable as a long term trend? And would this not be replicated when cheap Chinese labour is replaced by cheaper automated robots? I don’t have the answers, but I think its time people start thinking about them.


The impact of rates on banks

This post is from a mobile phone while I am getting ready to sleep, so may not make much sense. With this warning I shall dive right to my topic

Indian banks tend to have their stock price go up when the central bank lowers rates. For the longest time, I had never questioned this ‘fact’ and never stopped to wonder why.

When I did, at first the answer seemed obvious. Of course Indian banks prices would go up. After all the cost of funds is driven by the risk free rate and when the RBI drops it, the cost incurred by the banks comes down. What’s to question?

Then I thought about it some more. And this time, I put on my central bank hat. RBI (or any central bank) does not drop rates in order to help banks make more money. They do so in order to have more people take loans to grow the economy. The thing that banks should therefore do is to lower the rate at which they lend money and thereby keep their margins stable.

However, banks also have costs other than interest. They have to pay their CEO’s and CFO’s; their risk managers and their Vice Presidents who are responsible for sales. They need to pay for branch leases and ATM machines. Operational costs that are not interest. So they should actually make less money when rates go down.

Luckily there exists another driver. When rates go down more people start taking loans. This means that increased loan volumes make our CEO’s and VP’s work harder and therefore they get more productive. This leads to more profits and happy shareholders.

So the Indian example is now explained. What about places like the US or theEurozone? There the cost of money is very close to zero and any lowering of interest rates may not be of use in increasing lending.

In these cases, I suspect stock prices of these banks will go up when interest rates rise. We can wait and watch as the Fed is likely to raise rates over the next year and see if what I said turns out to be true.

Some bonus points for any readers who can identify the assumptions behind this post.