Why Your Doctor Is Blind (Statistically)

Medical-Crisis_quote-stats
Does your doctor understand the test results he has of you when he writes a prescription? Along with that does he understand the higher consequences of writing inaccurate prescription in his ignorance of interpreting the data incorrectly? We as normal people are supposed to rely on the experts for complicated matters. But what if the evidence suggests that on the face of increasing medical complexity and lack of statistical training it has become more arduous task. 
I have always got a kick out of finding flaws in doctors recommendations and their understanding of probability in their profession. It started when I went with my brother for treatment in Asian Hospital for frequently running nose. The doctor in the otorhinolaryngology (ENT) department after initial inspection told us to get the x-ray done which upon later examination convinced the doctor that operation is needed (by then I already had a sensation of the something wrong, I was amateur at this but sometimes I read extensively sometimes on the subject) to straighten the bent ‘septal cartilage’ (middle top of the nose). We checked with another doctor and he gave a kind of nasal spray and all was good within a month. That day we had a huge discussion in the family about the ills of medicine as a profession but I knew that something else was at play. I intended to find out and came to conclusion that: Medicine is a field of uncertainty and probability and it is a prerequisite for doctors to think in probability (which they clearly are incapable of) to justify their practice. Medical, it seems as profession is not too different from investment field.
   Experienced doctors begins the process of diagnosis upon first laying eyes on the patient and probability is what they use (although it can begin even before laying eyes). Based upon the expertise, he can expect the type of patient to come during the whole day and has some idea of the diagnosis he is about to prescribe. These starting-point likelihoods are called “anchor” probabilities (i.e, based on previous information). These probabilities gyrate based on subsequent history, examination and supplement testing (if necessary) according to what the patient has to say and what does or does not come to knowledge upon physical examination and testing. But strange as it may sound most doctors write more tests than what might have been justified only if he knows about the anchor probabilities. Apart from this there is another medical principle in diagnosis called ,’Bayes Theorem’. It states that the probability of a diagnosis after a new fact comes to light dep
ends on what its probability was before the new fact was
added. More simply, upon knowing the history of disease or physical examination or a dark spot on MRI testing might change the conclusion of a diagnosis from an earlier ‘yes’ to ‘no’. Each additional information means different in different context. Yet another implication of Bayes’ theorem is that one can’t skip past the history and examination by ordering a test in isolation and expect it to make an accurate diagnosis. A test is an answer to a question. If there was no question, how could test be an answer? Nevertheless, a doctor’s insistence on getting the tests done may mean that the doctor has ‘Skin-In-The-Game’ ( the idea as proposed by the author of Skin In The Game, just in a negative sense).
    In yet another case of amateurishness is with a little known simple statistical but intuitive concept known to non-medical population is ‘Number Needed to Treat’ (NNT) which even to medical people is a bit difficult to grasp. It is well known that with any type of medicine-some are befitted, some are harmed and some are unaffected. NNT shows the impact of medicine by estimating the number of persons that are to be treated in order to have one positive impact of the medicine. Say, if 1 patient
 is benefited out of every 2 patients treated the NNT is 2. Let’s take a hypothetical situation (real example is taken later on): The disease is fictional heart attack and the treatment we have is called ‘Medicine-1’. Imagine that 75% of heart attack victims who take Medicine-1 survive but only 25% survive
if they don’t take Medicine-1 (See figure 1 ). The upper 25% of the people will survive irrespective of treatment. Similarly, the lower 25% will die irrespective of the treatment. The people in the middle are the ones affected by Medicine-1 which in this case reduces the mortality rate by 50% (also called Absolute Risk Reduction or ARR). (Hint: NNT is reciprocal of ARR).
Medicine-1

Figure -1

The catch here is that we don’t know in which category the patient will fall into (benefited, not affected or harmed). So the ethics if the profession tells us to be realistic and estimate the likelihood of each one of the possibilities. The estimation works like this: The NNT of the medicine is 2 because the medicine affected 50% of lives. This means that “1 in 2 heart attack victims are affected by Medicine-1, or that “there’s a 50 percent chance that treatment with Medicine-1 will save a heart attack victim’s life.”
   Simple, huh? But wait the 2nd order consequences are even more fascinating. Let’s take another scenario in which 10% of people (out of 100% of population) die of heart attack if they don’t take Medicine-2 (in this scenario) and 8% die if they take Medicine-2. This is an ARR of 2% which means 2 out of every 100 are saved by using Medicine-2, for an NNT of 50 (=100/2). See figure 2. So, just like scenario 1, a significant number of people will be unaffected by the treatment (90% survive & 8% will die irrespective of treatment). Effectively, 98% of the people subjected to it are unaffected by it. Here’s comes the interesting part: the estimation of the 2nd scenario can also be stated as there is 20% relative reduction in risk (RRR) by taking Medicine-2.  The problem with this type of description is that, although it is semantically correct it is profoundly misleading. This is because before a treatment begins patients and doctors don’t know whether a patient will be helped, harmed, or unaffected by the treatment. If, in a conversation with a patient, we use the 20% to describe how likely it is that Medicine-2 will work (i.e. “Medicine-2 reduces your chance of dying by 20%”) then we have ignored the much greater possibility, 98% as we calculated above, that a patient will be unaffected. When we use the 20% to describe an effect of treatment we have concentrated only on the people who will die and ignored all of the people who will live. And we don’t know which group a patient will fall into until they have the treatment. Therefore using a description before the treatment that ignores the much larger chance that they will be in the group that survives regardless of the treatment (while still being subjected to the potential harms and side effects of the treatment) is very misleading. Actual picture will be like (Figure 3) an image that shows everyone (rather than just those may die).
Medicine-2

Figure-2

Medicine-3

Figure-3

One can see how using the RRR to describe the potential effect of a treatment would be enticing, particularly if someone wanted to exaggerate the potential benefit of a treatment. For Medicine-2 the RRR was 20% while the ARR was 2%, a ten-fold difference. Therefore people and groups that are trying to emphasize beneficial effects (groups that have a profit motive, doctors that are attempting to sway a patient in one direction, etc.) can use this logic.
The problem is that our brain has not been evolved to think in probabilistic terms. The coding of our DNA for the past millennia has been to hide and dodge the predators not to calculate, how much there is a possibility that a hungry lion staring at our face is likely to leave us in peace. We didn’t have the tools. But the irony of the situation is that we still couldn’t think probabilistically when we have the tools and can, even more so the professionals, leave alone the journalists, politicians and patients. A more surprising fact (less surprising by now, to be frank) is that most of the doctors don’t understand the concept of survival rates (its not opposite of Mortality rates – number of people die from a disease). I means the number of persons who have survived for a certain number of years (usually 5 years) after being diagnosed with a disease. In a survey of 412 doctors done in US by Gerd Gigerenzer, director of the Harding Center for Risk Literacy in Berlin, shows that 2/3 doctors mistakenly believed that higher survival rates means more lives are saved, which is wrong. The survival rates doesn’t count the patients who die from disease afterwards. The reason for this is we automatically think probability as a percentage. How? Here’s a question Gigerenzer asked to a group of more of 160 gynecologists: (give it a try and see how much you can make sense of it):
A 50-year-old woman, no symptoms, participates in routine mammography screening. She tests positive, is alarmed, and wants to know from you whether she has breast cancer for certain or what the chances are. Apart from the screening results, you know nothing else about this woman. How many women who test positive actually have breast cancer? What is the best answer? Options : (a) nine in 10, (b) eight in 10, (c) one in 10 & (d) one in 100.
He also supplied additional data women of this age group to help them answer:
  • The probability that a woman has breast cancer is 1% (“prevalence”).
  • If a woman has breast cancer, the probability that she tests positive is 90% (“sensitivity”).
  • If a woman does not have breast cancer, the probability that she nevertheless tests positive is 9% (“false alarm rate”)
More than 50% chose option (a) and 21% chose option (c) which is the correct answer. This is where Gigerenzer works focuses. He tries to make probability-oblivious doctors understand it through numbers. The fact that 90% of women with breast cancer tested positive (sensitivity”) from a mammogram doesn’t mean that 90% women have breast cancer. That’s an answer in ignorance of the other two additional information. When (“false alarm rate”) combined with (“prevalence”) roughly 9 out of 10 women don’t actually have the cancer even after tested positive (Figure-4). Since 1% is the prevalence rate (percentage of people having disease at a particular point of time) 10=1% of 1,000, out of which 9=90% of 10 are tested positive and 1 negative, there are still 89 women who doesn’t have cancer and will be going home with feeling of anxiety. Further, the research also suggests that months after a mammogram false alarm, up to a quarter of women are still affected by the process on a daily basis.
Cancer Test

   To quote Scot-Irish mathematical physicist Lord Kelvin, “If you cannot measure it, your knowledge is of meager and unsatisfactory. ” Anyone should be condoned in his ignorance of the fact but it would be more regretful of someone’s getting acquitted even after a mountain of evidence staring at his face against him, especially when a decision can have the profound consequences. The fact that more people are killed in US from medical errors than road accidents than plane crashes, terrorist attacks, and drug overdoses combined. And there’s collateral damage that can go unnoticed: Every day, doctors and nurses quietly live with those they have wounded or even killed, it would serve good for the society as a whole if it is made mandatory for the doctors to take elementary classes in medical-probability specially revising premedical education standards to incorporate training in statistics in favor of calculus, which is seldom used in clinical practice. Unsurprisingly, statistical probability is not taught perhaps because  many students in such programs already have an advanced science degree. But that doesn’t mean that they actually understand statistics. Yet another research had showed something called Dunning-Kruger effect among the doctors which means that doctors who weren’t good at statistics were more likely to say that they had good training in statistical literacy than those who did not. Doctors who didn’t knew statistics were so inadequate that they didn’t realize there was any more to know, whereas those who did know some statistics at least had a faint inkling that something was missing.
It’s not a problem of the medical mind. It’s a problem of training at the universities, in the medical departments where young doctors are trained in everything except statistical thinking.”– Gerd Gigerenzer
Similar to doctors, patients’ misconceptions about health risks are even further off the mark than doctors. Gigerenzer and his colleagues asked over 10,000 men and women across Europe about the benefits of PSA screening (prostate cancer test) and breast cancer screening respectively. Most overestimated the benefits, with respondents in the UK doing particularly badly – 99% of British men and 96% of British women overestimated the benefit of the tests. A quarter of British women went so far as to guess that 200 women out of every 1,000 screened have their lives saved by mammograms. But Gigerenzer says the real figure is about one woman in 1,000 – four out of every 1,000 screened women (Medicine-2) die from the disease, as opposed to five out of every 1,000 unscreened women (No Treatment). He says that this benefit has been represented as a “20% mortality reduction”, (20% of 5 is 1) which might explain why many women in the UK seem to think that 20% of women are saved by undergoing the procedure.
    In the face of increased risk of being sued for unintentional and unforeseen consequences majority (93% in a 2005 survey) of the doctors like to play ‘defensively’ (recommending treatments that are less likely to get them sued). In order to get them out of defensive behavior it would be useful if you were to ask questions in a mitigated speech because asking directly may infuriate the doctor (in which case it is better to switch the doctor given his disinclination which might change his behavior from defensive to aggressive) like,”I have been doing some research on my own and have stumbled upon some controversial facts. Can you guide me through some of them? “It’s surprising that in the 21st Century, many still think of doctors as Gods  and still they don’t ask the Gods” says Gerd. In medical ethics there’s a principle called ‘principle of informed consent’, which obligates the doctors to (a) disseminate the information which he knows about the diagnosis in a comprehensible way & (b) come clean about which he don’t know.
    “Have you done your body checkup? if not, Hurry! 30% discount on full body checkup till 31-May-2018.” Most probably you would go for similar type of tests (if you haven’t gone through it recently and cost is not a resisting factor) since it is recommended all over the place to get tested on a regular basis. If you ever come across these words like these never lull yourself into these misleading advertisements. Did I say ‘Misleading’? Before you start to loathe me let me make it clear that it exactly is what a recent research (and many others too) shows. The reason is that a large number of screening tests give false positives (the idea that a test comes out positive but the chances of having the disease is very less). Its mind boggling to think something like this exists and the doctors and patients have no inkling of this. In the research paper that described a test for ‘Mild Cognitive Impairment’ (MCI), a condition, that may, but often isn’t, a precursor of Alzheimer Disease (AD). The test was published in  the Journal of Neuropsychiatry and Clinical Neurosciences by Scharre et al in 2014. It had a Specificity of 95% (means 95% of people who are healthy will get correct diagnosis, i.e, tested negative) & Sensitivity of 80% ( means 80% of people who have MCI will get correct diagnosis,i.e, tested positive) & Prevalence of 1% (% of population that is affected with a particular disease at a given point of time). Using Gerd’s idea of thinking in images and numbers the calculation is shown below (Figure 5). 1% (100 people) will have MCI of which 80% will be tested positive and 20% will go undetected even when they have MCI. Similarly, out of 9,900 healthy people, 95% will be tested negative (correctly) and 5% positive (wrongly). Consequently, total number of people tested positive are 80+495=575 out of which 495 are false positives which is 86% of total. More simply, there is only 14% chance of actually having MCI if tested positive and 86% chance of not having it. To get the feel of this, think if a person sitting nervously in front of a doctor waiting for the results and the doctor says either of the two statements (a) there’s a 95% chance of you having it (doctor-fool) & (b) there’s only 14% of you having it (doctor-cool). One of them can make his day and another will wreck havoc on his family.

significance-screening-Fig-1

After having spent enough of my time understanding human folly believe me it gets frustrating some times thinking inadvertently of the errors (only probabilistically). Very recently I realized that more of my energy is used in trying to stop my impulsive observation of bêtises than volitional ignorance of it. Learning a little bit of statistics can avoid millions of unnecessary tests and deaths all world especially if concerns you and your dear ones life. Now that I’ve got the sensationalism, try this simple question yourself and test if your brain has understood what you have read :

If a test to detect a disease whose prevalence is 1 out of 1,000 has a false positive rate of 5 percent, what is the chance that a person found to have a positive result actually has the disease?


P.S:  If you couldn’t get the answer the first time, don’t worry. Of 61 physicians, hospital staff and medical students asked only 14 gave the correct answer — 2 percent.


  1. http://www.dcscience.net/2014/03/10/on-the-hazards-of-significance-testing-part-1-screening/
  2. http://www.dcscience.net/2014/03/24/on-the-hazards-of-significance-testing-part-2-the-false-discovery-rate-or-how-not-to-make-a-fool-of-yourself-with-p-values/
  3. http://www.cordingleyneurology.com/probabilitydiagnosis.html
  4. http://www.thennt.com/thennt-explained/
Advertisements

The Information Seduction

171130152515-free-press-information-overload-new-media-super-169

When I was in high school there was an exercise in which the whole school participated (not sort-of which gets you a memento). At the end of the lunch break three bells rang: the first one indicating the lunch time was over followed by 2nd sound at which whole school would stop in their position instantly like a statue. It was a hilarious moment to see, children making funny positions with their mates, others falling on the ground as if they were in the middle of the fight with funny faces. The seniors, didn’t caring about this would go on with their movements as if they are not afraid to break the rules (including me) or the reverse can also be said, they were afraid what others might think of them (social proof) if they would do the same. Well, the purpose of the exercise was to slow down our senses for the moment and take a look around and observe what is happening all around us. It was more of a mental exercise than a physical one. The world is moving so fast that we rarely pause for a moment and think what is happening and investment is no exception. That was 8 years ago. How all this connects to the investment in particular and life in general? Investment is like that, if you don’t understand the information you already have, you will be flooded with it by searching for more. 

It’s relevance has very much to do with the returns in the investment field. With all the marvels in the information technology that are supposed to make the information availability faster (speed) and with more ease than before (quantity) to rule out any insider advantage to the information and bring transparency, things have gotten much difficult in totality. We are more distracted than ever before with continuous bombardment of information. Simplification too, has a threshold limit. To borrow a phrase from Michael Bhaskar’s book Curation :  In fact, we have solved it (information scarcity) so well that there’s a new kind of problem: not information poverty, but information overload.

0_Am9pzF4sslCKmqaj

Gekko very clearly says it but didn’t say that the information he was talking about was curated information.

The logic behind superior investment results is simple, yet majority either doesn’t get hold of the idea or it is too incomprehensible for them. There can be innumerous other explanations for this but, that is not our concern here. Back to basics, superior investment results comes from information. Information has value in two ways: (1) Information that has not reached to many people and will have significant affect on the business (Insider information) and (2) Not everyone has access to the insider information, so what special insights we can develop from the information at our disposal. This article is about the second option by taking an example.

As it happens quantity matters but, to the extent it doesn’t lead to over analysis. It doesn’t require much empirical evidence to show that there is an inverse relationship between quantity and quality of information. More precisely said by the 19th century play-writer Douglas William Jerrold: Quality, not quantity is my measure. We all like to think that the more information we include in our analysis the better will be our decision for investment. Neurologically, information is addictive. Information is the most important commodity, but what if each extra information reduces rather than increases the quality of decision making. This question was raised by Stanford and Princeton university scientists in a fascinating study titled On the Pursuit and Misuse of Useless Information. Following is an observation from the study:


Decision makers often pursue noninstrumental information—information that appears relevant but, if simply available, would have no impact/less impact on choice.

The case explored here are motivated by the assumptions: (a) The pursuit of missing information can lend greater weight to that information, relative to the attention it might have received had it simply been known from the start; & (b) the longer the time to clear the uncertainty, higher the weightage is given to the missing information. These two factors are mutually enhancing.

The Curious Case of Vakrangee Ltd.

The company has been in news very recently for rumors of stock manipulation by the promoters group. Here is company that just before the crash (29-Jan-2018) was a total rage among investors. There were all sorts of media reports saying how the company has been a wealth creator for its investors over the last 10 years and how it’s earnings will double/triple over the years. Then, came the tide and everything was turned upside down. A market analyst was quoted as saying (after the alleged article came out):

“How the company makes money, God only knows. It is a case where promoter himself gives money to people to purchase shares of his own company. Perception of Vakrangee is bad in the market, but surprisingly, the stock prices are going up without any fundamentals and big names are seen in its shareholding list.”

All this was due to a article in a regional Mumbai newspaper claiming SEBI investigation into the company. Almost overnight the company went from wealth creator to  wealth destroyer falling as much as 70% within a span of 32 days. Brooding one’s mind over this fiasco should make anyone question: How is it that a near 50,000 crore company at its peak got reduced to 15,000 at its low within a period of 32 days? Did, its fundamentals change or the business was already in the sinkhole? The answer is NO! The reason is above two assumptions. As per the first assumption, the pursuit of missing information (some HNI’s even started their own private investigation) lead to downfall of the price which was more than substantiated. The information was relevant but non-essential for the price fall. During the uncertainty period more weight-age was given to a single information than all other information. I am no researcher but relying on the research, I can say that: had this information been available from the start, things would have been quite different. Try answering the following question:

Q: For sometime you have considered buying into this company but couldn’t due to high price. The company is good with great ROE, Sales CAGR & Profit CAGR for the last 10 years with reducing debt on the verge of becoming debt free with great future prospects. The price is down 50-70%. However, there are rumors in the news about some price manipulation in the price 3-4 years back by the promoters. You will not know until the end of the week whether the rumors are true of false. Do you

a)  Decide to buy the stock during the week?

b) Decide not to buy the stock during the week?

c) Wait until the end of the week, to decide whether or not to buy the the stock during the week?

If you choose (c) in the above question, answer the following:

It is the end of the week. You learn that the rumors are true and the promoters has been charged by SEBI. Do you

a) Decide to buy the stock?

b) Decide not to buy the stock?

I don’t have the resources to conduct the survey ( being a teacher would have helped) hence, I have no actual results for the questions. But I rest my answer on the rationale that, to any savvy investor it was clear enough that after 50-70% fall in prices based on a rumor which even if was proved right couldn’t justify the swing of the pendulum at the extreme pessimistic end, especially after knowing that there has been no disruption in the growth of the company as per the latest reports of the company (of course, unless you don’t believe in that too).

Bottom line: All else being equal, the probability of a stock going south of a good company by 50-70% within a short period is less (except the bad ones) and the probability of going it further down by 50-70% on a piece of information that turns out to be true (excepting the exceptions) at a later date is even less. As Howard Marks says: Any asset, no matter how low the quality is, can be cheap enough to be a good investment. How low the price should fall to be a good investment is a matter of personal judgement and second level thinking.

Hence, although the actual results of the survey might be different from what one would conclude by himself, it is clear enough going with option ‘a’ would have lead to a better investment results.

The second assumption relates to the time factor. Consider the following improvisation to the above case:

(a) Had the article about the company been based on actual information and the allegations were true right from the start or if the rumors were discarded in the beginning?  or

(b) Actual time it took for the uncertainty/rumor to dissipate.

Would the percentage of fall have been quite similar in both the cases to the actual fall. Well, the studies show opposite of that. The answer could have varied by 30-40% on either side.

Again, with the risk of being proved wrong, I believe that the fall wouldn’t have been so steep. The time it took to clear the rumor was enough for people to discuss and gossip among themselves enough that they started believing it to be true. Anyone with not much of expertise could have figured it out. To state that, this is not surprising, is to state the obvious. Some of the best ideas can come from special situations like this provided one has a grasp of the gravity of the situation.

Another aspect of the business model of Vakrangee is that, it is service based company and majority of the service is availed by the lower sections or middle class of the society that have no idea about stock manipulation (unless they happen to watch the business channels) and if they did somehow watched the news still won’t have much idea about what it means. A question that Warren Buffett used years ago when buying into American Express company, can be again repeated here: Does people mind/stop using services of the company after this. Till people are getting the best of services, they don’t care much about any scandal in the company. They are too busy in their life to stop and think about some mischief done a promoter of the company.


Conclusion: The research underscores a sobering message: We’re fascinated with filling information gaps and that obsession can lead us astray. Especially today, when reducing uncertainty has become all too easy. Getting something meaningful out of the information at our disposal is the first step towards investment. In a world where every click brings the promise of a discovery, we are all at risk of becoming information addicts. The challenge lies in differentiating between questions worth exploring and questions best left unasked.


Disclaimer: The author holds shares in the above mentioned company and hence, the opinions expressed might be biased. Do your own homework, before coming to any conclusion. This is not meant to be a solicitation or recommendation of stocks.


 P.S: I want to leave with one question as a thought: PNB Bank has been hovering around its 52 Week low for few weeks. Consider the following facts about the bank and then answer the following questions: PNB is the (as on 23 March, 2018):

  1. 2nd largest bank in India by total assets
  2. 3rd largest bank by Market Capitalization and Net Sales
  3. 4th largest bank by Net Profit and Cash and Bank Balance and EPS
Question: (a) How much more lower it can go? & (b) Is the probability of the stock going lower more than that of going higher in the next one year? 


For more articles like this follow my blog and/or follow me on Twitter.


The Debt or Death Clock

National Debt

                                                          Source: http://www.commonsenseevaluation.com

There is an interesting website National Debt Clock which shows the world debt on real time basis and interest accumulating on it. Data for individual countries is present too. Clicking on India will show interesting facts about Indian debt. As of writing this $69.7 trillion dollar is the world debt figure. Just like time goes in one direction, so does the total debt of the world. The real question is, will it stop? If yes, when? If no, then one thing is certain, the next crisis. Then the only question remains is, when will it occur? The clock is ticking. Every second, it seems, someone in the world takes on more debt. 

Does it matter? After all, world governments owe the money to their own citizens, not to the Martians. But the rising total is important for two reasons. First, when debt rises faster than economic output (as it has been doing in recent years), higher government debt implies more state interference in the economy and higher taxes in the future. Second, debt must be rolled over at regular intervals. This creates a recurring popularity test for individual governments, rather as reality TV show contestants face a public phone vote every week. Fail that vote, as various euro-zone governments have done, and the country (and its neighbours) can be plunged into crisis. (Source: Economist.com)

Capital is at the heart of modern economy. Virtually every growth metric in the world is somehow dependent on this measure of growth (GDP of a country). However, this growth has also lead to some disasters felt few times (more often than few times) in the history and that experience is now what calls for the caution when everything gets to extreme. Consequently, where does that capital come from and where does go is a matter of grave concern.

There are few references in life so common as that to the lessons of history. Those who know it are not doomed to repeat it.

Perhaps the greatest gift to the financial field is the fractional-reserve-system, wherein banks can keep a fraction of the original sum deposited and lend the rest of it. That has a multiplier effect. For simplicity, if 100 rupees is deposited in a bank account and 10% is reserve mandate then, the money multiplier would be =100/ reserve ratio, i.e. 10 times. So, 1000 rupees is created out of 100 intial deposit into the system. Extra 900 is nothing but credit granted to the borrower out of thin air.

Economist Hyman Minsky argues that excessive debt is the reason for most of the financial crisis in the history of the world from Tulip Mania (1637) to South Sea (1720) to Great Depression (1929) to Sub Prime (2008). Debt has left deep marks in our history of the world. According to him there are three types of debt.

  1. Hedge Financing: This is the safest debt of all. This is the primary reason for giving credit. The borrower invests capital in productive economic activities, from which cash flows are generated which pays interest and ultimately the principal. This was how the system was meant to work, at least at the start. Everything goes good till here.
  2. Speculative Financing:  Although the first stage is the most obvious but with a drawback. The speed of growth is slow. To its rescue comes the speculative financing. Under this the credit growth is achieved often through government by making policies that makes taking on debt less costlier through subsidies and priority-lending schemes. This is often achieved through relaxing lending norms. More people are now can under the umbrella of debt. More lending means more spending means more growth.
  3. Ponzi Financing: This is the most dangerous of all because borrowers use the capital not to invest in productive assets, but to buy assets in hope of selling it at a higher price in the future, repay the debt and book a profit. A good example of this comes from my personal experience. A friend of mine suggested a mechanism to make money very easily. His proposition was: Take a 5,00,000 personal loan from the bank with the cheapest interest rate (SBI February rate 12.55%-17.65%)for tenure ranging from 1-5 years. Assuming 20% is the minimum average return he can make and average rate of interest is 15% applicable, he will have to make money. His idea was good or bad is a matter of seperate discussion. But it gives us a pretty good idea when surplus capital is spent in frenzy.

Bottom line: He is in the third category of debt financing with the ultimate result being a disaster.

Minsky also concluded that long periods of stable economic growth where everything good happens leads to gung-ho situation on a wider scale and this fuels speculative and ponzi financing to dominate. This leads to inflation in prices, especially of stocks and real estate. Ultimately, when prices stop rising, debt defaults goes up and bubble pops and the economy goes down with it like house of cards.

An example is the 1997 Asian crisis that destroyed wealth across many countries including Thailand, South Korea, Indonesia, Malaysia, Singapore, Hong Kong, Taiwan and, to some extent Japan. These countries saw rapid economic growth in 1980’s and 1990’s, which lead to higher risk taking. Government banks started giving risky loans with less stringent repayment provisions to less reponsible borrowers. When there is excess capital it finds it’s way to the financial assets creating inflated asset prices. When asset prices stopped rising borrowers started defaulting on their loans. Foreign investors began to flee. This put pressure on local currencies. Thailand’s baht was the first to crumble, and as the dominoes started to crumble, panic spread. In 2012 Japan made headlines when the government’s debt reached 997 trillion yen, that’s 200% of GDP which turns to be $80,000 per capita. Similar to this was the 2008 crisis when $20 trillion was wiped out within a few days.

The drop in real estate prices has in a way exposed the Indian banking sector ways of lending. As per the RBI data published in December last year total NPA stood at 7.34 lakh crore of which PSB’s account for 87%. The NPA of SBI alone is more than the combined total NPA of private banks. With yields rising in USA, foreign investors will flee and RBI will be left with no choice but to let the rupee tumble- as it did in 2013 when the rupee dropped from 54 to 68. All that the central bank can do is increase the interest rate which is not a viable option when the banks are already reeling with the NPA issue.

The inflated asset prices in US, negative interest rate in EU, the looming crisis in China and NPA in India is something more than a short term problem at hand. Without the discipline imposed by risk based lending, excess supply creates asset bubbles. The Minsky moment is looming larger than ever on India  and China, with Debt to GDP ratio of China reaching 260% of GDP. Historically, any country’s ratio crossing 200% is a sign of future crisis in that country. The next financial Crisis may very well start in Asia more specifically China. Below is a snapshot of Debt to GDP ratio of India.

statistic_id271319_national-debt-of-india-in-relation-to-gross-domestic-product--gdp--2022

Bottom Line: Debt is like double edged sword which we know by experience where way it will fall. We better prevent it from falling before it takes lives in the next financial fiasco.

Further reading: How India’s Debt Could Kill Its Growth

Long Is The New Short

On Stock Market

(source: Dilbert.com by Scott Adams)

It has always amazed me to see that how people can easily fall for the same error that they seem to correct at other times. The current writing is a consequence of an article I read recently in the last issue of DSIJ (Dalal Street Investment Journal) were in the editor-in-chief V.P. Padode said:

“Markets are poised in a such a way at this juncture that indexes may move sideways but there could be a lot of stock specific actions. We expect indexes to be in sideways trend for the coming couple of months at least while maintaining our targets for Sensex at 36000 levels by March 2018.

Giving targets works like placebo to the people to make them believe that they can measure the temperature with their tongues out even though scientifically that’s not possible. This reminded me of newspaper articles few years back making rounds at that time. I will go in ascending order year wise to make others understand how at different times widely followed and interviewed fund managers and CEO’s said and what happened in reality (or will happen). Taking a look at their predictions and analysing it will give us some answer to the predictive abilities of financial gurus even though there is none.

This news appeared in Business Standard on 5th June, 2014 (results of the general elections were declared on 16th May, 2014 in which BJP government won with wide margin) where Varun Goel, head of PMS, Karvy Stock Broking (there is something peculiar about names like these which makes us believe that the person behind the table must be knowing something, when in reality all he does is making people believe in something of which he isn’t sure of himself) predicted that Sensex could touch 100,000 by 2020. Another article on the same day in the same newspaper appeared wherein five reasons were given (I didn’t read them, not because I was overconfident that they would be worthless but because I was confident that, it just couldn’t happen)  as to why Sensex can touch 100,000 by 2020.

Here’s another one: On October 5th, 2015 an article appeared on Money Control  in which Utpal Seth, CEO of Rare Enterprises mentioned that Sensex can touch 50,000 by 2020. Many more can be mentioned but considering the effort and time, it would be a pointless exercise to mention them all.

Now let’s check what has happened since then and how much of that seemed plausible with the help of elementary mathematics in hindsight.

On the day of news article (first of the above) Sensex was at 25,019. To touch 100,000 within next five years, following needs to happen:

The market capitalization has to increase by 3.0517 times the market capitalization for the year 2015. The expert making this futile prediction also said that 3x is not very unlikely given that from 6,602.69 (2004) it reached 20,286.99 (2007) and went down to 9,647.31 (2009), then again touched 20,509.09 (2010) and reached 25,000 up until 2015.

dsdfhdsjf

Market capitalization to GDP ratio is one the most tracked ratios to find out the valuation of the market. As a general rule a result of greater than 100% is said to show that the market is overvalued, while a value of around 50% is said to show undervaluation, not to mention that determining what percentage level is accurate in showing undervaluation and over valuation has been hotly debated. As per the projected valuations of 2015 the ratio came at 1.028 times of 2015 data. Below is the trend of the past:

chart1498457279769

For those who said that its very much possible that mark of 100,000 can be achieved didn’t have a good look at this chart or they were the victim of self-denial. They didn’t understood that the economy of a country doesn’t increase at such a rate that the market would increase 3 times within five years. We can see at the past trend to see what happens when it does. Looking at the chart it becomes clear that from 2004 until late 2007 was the phase of extreme bull period buoyant by the cheer of housing prices in the USA and (in the not-so-extreme sense) 10 years was over since the the economic reforms and it started to bear fruits ten years after (yes, reforms like that takes this much time to show its signs on economy), so the bull of privatisation that was unleashed in 1991 was at its full speed after 2000 (Sensex reached 20,286.99 from 3,972.12 within a period of 7 years, 5x within 7 years). After the fall of 2009, it more than doubled the next year and for the next five year it increased by 1000 points per year.

The bottom line is that: For Sensex to reach 100,000 by 2020 it needs to grow by 25% CAGR per annum. The so called financial gurus prophecies were/are wrong for two reasons:

  1. The data they were referring to were picked at the wrong point in the history’s time line. A period when everything seemed good and nothing could go wrong (remember Reliance Power, the big one).
  2. For the financial market to grow at 25% per annum, all the other metrics of the economy (GDP growth, Credit, Savings, Per Capita Income, Employment, etc…) also has to grow around that rate to support the financial system. Because stock market is like the mirror of the economy, it depends on economy not the otherwise. Think about it, when was the last time our economy grew at that rate (China did it at that rate for at least a quarter of century and see where it is, China’s GDP for 2015 was $10.87 trillion, around 10 times of India). The answer becomes clear when we ask very plausible question, can India become a $10 trillion economy in the next five years? Is it able to grow at the rate which can justify a market capitalization of 100,000 points? The answer is no.
  3. An increase in index 3-4 times within few years (100,000/25,000=4) is exactly the thing that takes the market at the extreme end of the pendulum and it’s reversion causes people to loose their patience.
  4. An analogy: An economy is like the company and indexes (BSE/NSE) are its price. There’s a fundamental rule of investment i.e, a security cannot be worth more than the asset underlying that security, the spread between the price and value will converge over a long period. The increased indexes cannot remain sustainable if the growth rate of the economy isn’t on par with the indexes.

The problem with people is that they make analysis based on the long history of the company, forecasting it long into the future, then taking short term decisions  based on price gyrations…..As Howard Marks said “ The problem is not of informational or analytical but psychological.”

The Present:

As of today Sensex is at the peak of 31,138 and Nifty at 9,574 both of which are at their near highs. If the discussion going round the news is to be believed, both of the indexes are touching the clouds and there is a paranoia in the market that it is nearing its correction while some believe that still some gas is left to push it further near 36,000 or 38 or 40. I never understood hoe people come up with numbers like this. I mean one can only give a range of something like that but not a specific number.

Before making any assumption about the crest and trough of the market we need to understand the reasons behind it. Much of the people rely on their instinct or media clutter on which I have little faith.

In my opinion the markets are rich but not extended to such a point where a correction or crash is eminent. I can’t say where the market is headed (up/down) but, I can come up with a satisfying conclusion for no-fall-yet in market based on some some numbers:

  1. For the past few months, on an average 4000+ crores of money are flowing in SIP’s every month. That’s approximately 48,000 crores.
  2. Since EPFO was given the liberty to invest in the equity markets more than 10,000 crores have been invested by it and there is a plan to invest 18,000 in the current fiscal adding fuel to the stock market that is already on fire.
  3. Increase in NPS schemes is also adding around 2,000 crores a year.
  4. At last, the retail investor who were siding on the sidelines till now will also jump on the bandwagon to get the benefit of rising fortunes. Not to forget that more people will come to senses that stock holding is better way to stay wisely wealthy.
  5. All of the above turns out around 70,000 crores in the current fiscal. Amount of this much when poured in the markets will eventually push the markets in the north.

At the end the only thing that can increase the chances of being correct is that; no matter what or where the stock market is, the only thing that matters is that you have purchased an undervalued company at such low price that it will go up in with time. The fall in market shouldn’t be the concern of investors and when it does, only to the point that it gives us more opportunity to buy of what we already know is good at reduced prices.

Perils to Understanding

” A learned block head is a greater block head than an ignorant one” – Benjamin Franklin

These words of Ben Franklin echoes with greater force when it comes to making people understand the investing ‘process’ (emphasis is needed on the word). People are trained all through their schools that everything can be reduced to the extent that it can be applied by everyone and that’s the wrong way to think. You cannot reduce everything to some formula. I have been trying to make people understand (I gave few classes to students) that there is no standardized ‘process’ to investing. Its an art that needs perfection every time and the way to do it is to realize in the first place that there is no formula for success in investing. Then the most plausible question that they ask is “How can not understand something it you haven’t done it?” The way I explain it to them is that you don’t have to do operation in a lab to know that you cant do it. Here is a comic strip which best tells more incisively the mindset of today’s educational system be it any field:

CH940127_JPG

Prediction Begets Prediction:

Investors try to emulate the big investors of all time trying to predict or making speculations about what they are buying or selling. They try to know the exact procedure of how they go about making investment. The banality of the situation is that the very same celebrated investors in their investing carriers have denounced predictive value of such future prediction and yet people go on predicting what they are doing. Chris Davis of Davis Funds explains in this video in a very intelligible way about why the prediction business goes on and on: Davis Funds.

What people don’t get to understand is that no one can become the person that they are following. Thinking like Einstein you won’t become Einstein, you become only the best of who you are. We can read anyone’s investment philosophy about how they did this and that, but that wouldn’t be our own. The key to understanding the markets is to have your own philosophy of investment and that will come from what we have learned through our parents and teachers and what our experiences teach us about what we have been taught by our parents and teachers.  The bottom line is that there can be only one Warren Buffet, only one Howard Marks and one Peter Lynch and only one Ray Dalio and they have already walked the earth. There is not going to be another of anyone of them. Also all of them were unique in their own way. The world will continue to produce someone with great record like them but they wouldn’t be the same as these people. What we cannot do what they did nor can anyone else, on the contrary what we can do is to learn from them these great  people is how to be successful at what we do, not them and that’s what they have been teaching all their life.

Only when we understand this, that we will stop asking all the pointless questions and start discussing meaningful questions that not everyone asks every time. This will finding stocks where no valuable investment is made.

The following is a modified version of one of the best books by Peter Thiel “Zero to One” that best describes the point of this blog:

The next Warren Buffet won’t be running another Berkshire Hathaway. The next Peter Lynch won’t be picking stocks in Magellan Fund. If you are copying these guys, you aren’t learning from them.