Finance, Uncategorized

Recap — The 2008 Financial Crisis

Its been some time since I have posted on the blog, so I thought I would gently sidle back with a bit of a flashback.

The Financial Crisis of 2008 was an my first personal exposure to the effects of economic policy on real things like the price of vegetables. Until then, I was either too immature and young to even follow stock markets and the interaction with the so called “real items”, or I was an armchair observer, merely observing shifts in various underlying items without being able to correlate these shifts with an understanding of what these shifts resulted in.

But the crisis of 07-09 was quite a difference. It starts off when I was still in B School, having received along with a friend an almost 6 figure sum as prize money for a Business Plan competition. Of course, while father and family were already making house renovation plans with the money, my friend and I were busy asking R, our resident expert on everything “Whats the stock market”.

2008 was the year of arguably the biggest hyped listing in India, Reliance Power Limited or RPower. It was a big enough deal that everyone from my night canteen shop owner to the laundry man at the hostel applied for the IPO. Luckily for my friend (SA) and I, neither of us put money into that boondoggle. We found far better (and faster) ways to lose our money.

RV enlightened us about the wonders of the futures and options market, where leverage is (almost) unlimited and you can multiply your returns faster than anything. Being a smart cookie though, he demurred at the thought of actually putting any funds into this particular venture. His goal was higher than mere filthy lucre. At the time RA was trying to become the second largest shareholder of Bharat Heavy Electricals Limited (BHEL), and diverting funds to less productive tasks was not up his alley.

So SA and I decided to forge on our lonesome. SA had already acquired an ICICI direct account from his misspent youth at Infosys Chennai, and we promptly dumped a large portion of our funds into this account going long on the NIFTY 50. Cut a long story short, we wiped out our capital in 3 trading days.

And it was just my luck that during placement season, I was placed in the banking industry with a new private sector bank. What I knew about banking then was that banks took deposits and then put those deposits into loans. Deposit interest was less than the interest the bank collected on loans. So banks made money. Quite simple.

Well, I was disabused almost immediately. I was sent to the corner of the banking world called trade operations in 2008 May. This is the place where they issue bank guarantees and Letters of Credit and do things like Buyers Credit. Not a loan to be seen no matter how hard I looked. The local Bank Guarantee market was not really affected by the financial crisis. But the Buyers credit market increasingly began looking more and more tense as the year went by. By the time August and September rolled by, the credit crisis was not some global event that I read in the headlines, but a daily grind of what new craziness LIBOR was up to. Looking back with the benefit of 10 years of hindsight, it seems quite funny, but I assure you that it was not. As a newbie in the industry with no baseline to compare against, I thought that this was the “normal” day in the office.

And we were a new bank with relatively low legacy book that was going bad in our face. Octobor came and went and markets tanked like they were never going to recover again. That is when SA and I put a small sum of money into stocks which we thought would not go out of business (ICICI Bank, Reliance Industries, Infosys are the ones I remember). By March, our investments were under water and my bank’s stock price had tanked to about its issue price during its IPO. And my workday looked like purgatory, heading into office by 8:30 am, and barely catching the 11:30 pm local train back home every night. But like most things, it was darkest before dawn. By April 2009, things were looking up, and the world was a much more stable (albeit less interesting place).

As a postscript, me of a decade ago thought nothing of coming home at midnight and then tracking US markets till their close and then going to sleep and still be able to wake up and work the next day. Now, the very thought of doing that for 6 months at a stretch makes me want to go and take a nap!

Economics, Uncategorized

The Inflation Mystery — Part 2 — The components of CPI

Yesterday’s post was a light and easy introduction to the idea of QE. We did QE at quite a high level because we don’t really need to get into the weeds about which bonds need to be bought, or for how long. The background that I wanted to provide was that we are in an easy money world where capital is relatively abundant.

But this post is not going to be like that. Oh No. If you want to understand why this creature called inflation has become so notoriously hard to find, its time to head out to the tall grass! Like hunters chasing after particularly vexing prey,  we need to search for signs of its passage. There will be lots of piles of steaming turds we need to examine to evaluate the path of travel of our target. So let us dive in.

First up though, a warning: Statistics is not my specialization. And I don’t want to wade through equations where a graph or a chart will do the job. However, interpretation of data is a skill that you, dear reader certainly possess. And if you do not, I urge you to grab the opportunity to learn along with me! So, with this warning done, onward fellow hunter.

Developed world Inflation components

The minions who scurry around in the US to provide the inflation data we get have to first identify what things prices are to be tracked, and when tracked, what weight should we attach to them. It would be easy to track inflation by a single metric (See the McDonalds happy meal as an example), but it doesnt tell you as much as a broader indicator that more accurately represents how the median person spent their money.

Broadly, we can split an economy into goods and services. The US has a split as shown in the graph below


Fig 1 — Split between goods and services

As you can see, services moderately trumps goods in the inflation weights. Its almost 6:4. But the more interesting bit comes up going forward. If we drill down a bit more, the numbers show some more detail.


Fig 2 — Service sector breakup

The services part of the inflation data shows one very large component. That is shelter (proxy for housing). Now those who rent houses will nod in agreement and say, “that makes sense”. But for those who do not, and live in their own house, it is a bit of an odd one.  For those who own houses, what is calculated is the imputed value of rent they would be paying themselves. This seems to be quite sensible in theory, but will make inflation numbers quite subjective for people who own their own houses.

The other major constituents are Medical services, transportation services and education. Alongside this, recreation and the miscellaneous items do count, but not for as much.

Now lets have a gander at the goods side of the equation


Fig 3 — Goods Sector Breakup

In goods as well, there are 2 big constituents. First is food, which seems enormous. Then comes energy, which is a fancy way of saying electricity and petrol bills. Apart from that there is buying of cars and spares, which makes me suspicious about whether I have made this chart correctly! But you can see that goods inflation gives about 50% to food and fuel.

So mixing the two charts together we get some pretty interesting outcomes.

Food First

Food, transportation (cars, fuel and services) along with shelter make up well over half of the basket of inflation (about 70%). For inflation to go up, these items need to move up. But, these are relatively sticky. Most people eat the same stuff they always do. They may splurge a bit by eating out some more, but it cant go up too much given the current data on how much the average American eats out! (For the curious, one in every 6 meals is eaten out).

Still, food prices can go up, can’t they? Not really. Food prices don’t double in the developed world week on week. The volatility is relatively small. Given that quantity does not change much, nor does quality of what you eat is relatively constant, that takes out food from the equation.

The data I used to base my conclusion is shamelessly stolen from here. I urge a visit to look at how the inflation value over a very long period has been trending down.


Well, transport and fuel are a very large component indeed. But lets have some flashbacks to some bad old days. Oil used to be 120 dollars a barrel in 2007-08. Today they are half what they were. (Even if double the price from the lows). Most commodities share the same fate. So the cost of buying cars and fuel and getting from place A to B have not changed much. The long term trends do indicate an increased distance every year since 1990, but not enough to offset the lower prices. (You will have to take my word for this).


Last comes housing. Now the US has pockets where the housing sector is certainly super hot (Hello San Francisco!). This is the only section in the big 3 of the inflation basket that actually increases at the Federal Reserve target rate of 2% (it is slightly higher). But even with the large weight, its relatively sticky. Housing inflation of 3% is imaginable. But in the absence of very large changes in income or population (or percentage of home ownership), rental costs are unlikely to change significantly beyond the trend rate.

The fastest growing measures of inflation are healthcare and education. They cause a large amount of noise among those affected (as they should), but as of today, they still make up only a relatively small proportion of the overall pie.For the present, they don’t do much more than round up the inflation number slightly. It seems very probable that healthcare at least will become a much larger component going forward though.

So the mystery has sort of been solved. The inflation has not vanished as such. It has been hidden amidst the weeds of food and transportation, and the housing crisis had made it hide in the basement for some time. But for those who stock wheelbarrows just in case they need it to cart currency for those loaves of bread, a look at the trends shows that it need not come out of storage anytime in the near future.

But this tale is not yet done. We will take a look at some of the assumptions that may change the inflation tale. And we still need to take a world tour through the developing world and Indian data! That will be for another day.


Economics, Uncategorized

The Inflation Mystery — A developed world saga

Prologue — A primer on Quantitative Easing (QE)

Well, I needed a break from ISDA reading, and I found it while walking down the road a few days ago. Inflation — Financial pundits of a particular ilk were united in their criticism of  the ultra low interest rate policy followed by the US Fed from 2008 onwards.  They were adamant that with super low interest rates, people would immediately clog markets and start purchasing goods (of which there would not be too many). This would cause producers to hike prices. Knowing that their money is rapidly failing to buy as much stuff as before, consumers would throng to the stores in even greater numbers,  causing producers to continue hiking prices. Then this vicious cycle would repeat, until we would all be carting wheelbarrows full of money to buy bread, and the world as we know it would disappear in a puff of monetary smoke.

But what do you know, that dire fate did not occur. Not only did the market for wheelbarrows collapse, inflation stubbornly refused to tick up. As a matter of fact, around the world, the fear was that of deflation, or prices going down. Banks went all out the other direction now. The US Fed debuted its next tool, called quantitative easing.

Before we start speculating about why wheel barrows are not needed and all that, lets start out with a primer to explain QE

Quantitative Easing (“QE”):

So what is QE? 

Simply put, its buying long term government (or other) bonds from the open market. This increases the prices of those bonds. Since bonds are weird, the rise in price of the bond automatically lowers the interest rate that you receive from these bonds (yield). By lowering long term yields, cost of long term borrowing comes down. This is supposed to make businesses more certain of interest costs over a long period of time. And so they will commit to building capacity and employing people. And these employed people will go out and buy more things. Which will make producers buy things and then presto! Inflation!

So whats the difference between QE and standard monetary policy?

Not as much as you think. Standard expansionary monetary policy buys short term bonds, thereby dropping overnight interest rates. But its pretty hard to do this when rates approach zero (don’t tell the European Central Bank this). So QE is the tool that the US Fed used once rates came close to zero.

Wait…What was that crack about the ECB?

Well, unlike the conservative Americans, Europeans know how to let down their hair and party. So when overnight interest rates hit zero, they said “who said that interest rates have to stay at zero. Let us see if you can go lower than zero!” And negative interest bonds were born! To this day, I see no plausible use of these instruments for an actual investor! However, there are 2 ways you could conceivably make money if you are a trader.

Way number 1: The greater fool — Here, you buy the bonds that give you negative interest in the hope that the price of the bond will go up, so you can sell the bonds to the next sucker, and then not have to ‘receive negative interest’. So you make capital gains on the bond

Way number 2: The currency dude — Here you borrow money in USD at interest rates of 1% or so, and then use the borrowed money to buy -0.25% EUR bonds. If the USD weakens against the dollar by 2% in a year, when you convert your EUR back to dollars, you have 2% more dollars. you give 0.27% of this to the Europeans (Negative rates), and 1% to your lending institution, and keep 0.73% as your profit.

So how does QE work?

Well, this seems like just another way to lower interest rates. Only difference being you target longer term rates rather than purely overnight rates. ‘What’s the big deal?’, you may ask.

Well it is a big deal for a few reasons. One is that by lowering long term yields by buying bonds, you flatten the yield curve.

The Yield curve

The figure above illustrates this. The curves are hypothetical yields that US government bonds would provide over different maturities. The values are not totally wrong, but they are completely imaginary, so take that where you will.

Like any graph, it does more than just show the difference in yields. A closer look at it also shows the strengths and weaknesses of this policy.

QE has the potential to move long term rates. It is pretty much the only policy tool that central bankers have to move long term funding rates. And to put it bluntly, most businesses don’t really care too much about what overnight rates are. You are not going to decide to build a 2 billion dollar chip plant based on what overnight rates are. But if you know that funding costs for a 10 year loan are 10% lower than last year, and will remain low for the next 5 years at least, your decision to spend that 2 billion dollars becomes a lot (10%) easier. So QE works by making businesses that were on the “will we- won’t we” phase make the decision to will, rather than won’t. The other constituency that will care about lower funding costs are individuals who take long term liabilities like houses. It makes it a lot easier to buy your house at 3% than at 7%.

But that also is the weakness. A business may not care too much about 10% lower financing costs if there is no one buying the chips its making off their existing plants. The only guys you will help with this lowering of yields are existing loan takers. Mind you, this is not bad if there are a lot of existing loans that would be underwater at 7% that can afford to pay 4. But once the pool of high cost loans vanishes from the market, the QE strategy may not work very well.

And there is another weakness that will follow on from the low yield regime. There is a good chance that companies would use those low yields to load up on low cost debt, and use that low cost debt to do unproductive things. Just as an example, a solid gold fountain that sprays $1000 dollar champagne. I am absolutely sure no company would do such a silly thing, but it makes an awfully nice (albeit tasteless) office party! And its a lot easier to sell to your boss at 0.05% financing than at 5%!

Balance sheets and other shibboleths

Those financial pundits I mentioned at the start of the post also worry about things like ballooning Fed reserve balance sheets. They also make dire predictions of how governments get emboldened to do silly things with low cost funds. These are somewhat pertinent points, but they are will make this post quite a bit longer (and a lot more boring), so I will not tax your patience, dear reader.

I shall cover the main mystery. What happened to the inflation all those central banks were targeting? It will involve fairy creatures, travels around the world, and iphones. But all this will need another day, and another post to be explained. Till then, you will remain in suspense.

Derivatives, Tutorial

Derivatives Documentation — Post 3

In the last few posts we ran through what OTC derivatives are, and then wandered into 3 types of documents that are executed to enable one to make a trade.

Today we start off with the actual detail. Lets see how the ISDA Master Agreement(“MA“) is structured, and start off with the first few sections.

Lets imagine the MA as a pizza. Every pizza needs a base. In the same was every MA that is signed would have a base. the number of these bases is quite limited. Like you have a whole wheat thin crust, or a cheese filled base, the MA base is one of either the 1992 or the 2002 Master Agreements. Yes. there are older bases, but those are too old fashioned for us to go through.

Like pizza making, the choice of the base is determined by each persons taste. But its widely agreed that the 2002 agreement is a more robust agreement. This is primarily because it incorporated the changes that ISDA learnt through the hard knocks school of the 1997 crisis. Strangely enough, the 2008-09 financial crisis has not provided us a changed document, but that could also be because the stress test off the 2002 Agreement by the Lehman bankruptcy showed that the existing document worked mostly fine.

The problem with financial companies though is that they are old entities that have a whole bunch of legacy systems and relationships. So many banks and financial companies have existing relationships with financial companies that have used the 1992 ISDA format and finds it works well enough. The hassle of negotiating a whole new set of documents and migration can sometimes be too much work, so they may just go with the flow and sign new 1992 agreements. However, if you are a new counterparty starting a relationship with a financial institution, a 2002 Agreement is much preferred.

Structure of the Agreement

Post 3

The ISDA Master agreement is also called the boilerplate agreement. Its the text that does not change. All that changes between different agreements is the schedule to the MA. This schedule will enable/disable individual sections of the boilerplate. This is the meat of the document, as it customises the documents for two different counterparties.


This is best illustrated with the help of an example. The same ISDA master agreement is used for an agreement between a bank and a corporate as one between a bank and a hedge fund.

However, a corporate is an entirely different beast from a fund that manages money. For example, you would ideally not want to do business with a fund when it suddenly loses subscribers  (even if its performance is great). So you may have a minimum total assets under management trigger built into your ISDA MA. Once the assets drop below a pre-agreed minimum, the bank would have the option to terminate the agreement, and all outstanding transactions done under this agreement.

This is not relevant for most corporates. What you care about in the corporate case is that the majority ownership of the corporate stays the same. A sudden change in ownership and management would be a cause of sufficient concern that the bank would want to have the ability to close existing transactions and terminate the agreement.

These sort of clauses are inserted in the schedule to the master agreement. The boilerplate does not have these custom clauses built in, but it does allow for these to be inserted into the document. We will learn more about these sort of Additional Termination Events (ATE‘s) in later posts

The 2002 ISDA MA has 14 sections. Let the number not intimidate. Today we start light, with section 1, which is an easy one. I do not wish to offend the copyright gods, so i will once again urge you to the relevant SEC link, and will stick to the interpretation given below.

Section 1: Interpretation 

a) Definitions: All definitions in the document are labelled as such and will have the meaning that is given within the document. Simply put, dont rely on your understanding for words that are highlighted in bold or capitalised. These are defined terms, and the definition will be found somewhere in the document (usually in Section 14 of the MA, and scattered throughout the Schedule like confetti)

b) Consistency: Basically, The Schedule overrides any text in the Master Agreement. And the Transaction Confirmation that specifies the terms of the individual transaction overrides both. Just reverse the order of the helpful figure above, and you have the order of which document is the authoritative one. Of course, modifying terms of the ISDA in a transaction confirmation is very frowned upon, except for collateral terms (I am cheating a bit here. ISDA’s do not have Collateral specified, but they are usually signed along with something called a Credit Support Annex (“CSA“).

c) Single Agreement: This clause states that the ISDA, Schedule, any amendments, and the transaction documents all form a single Agreement. Seems like a simple thing. But this can actually be a very big deal. Remember, one ISDA can have many transactions under it. And each of these may be having a different value on any given day. If each transaction were independent, every transaction would then have its own Profit or loss. These values can be quite large. However, if you take all of them as a single agreement, you can take a total Profit/loss of a counterparty. This total would usually be a LOT smaller than the absolute values. And this matters a great deal for credit exposure. Again, an example will explain.

I have 2 transactions with another CP, lets call it “Oil Bank”. In one, for every dollar NYMEX crude is above 60, Oil Bank will pay me 0.5 million dollars. IN another, for every dollar over 65, Oil bank will receive 0.75 million dollars. For a given price of NYMEX Crude (say $67), the total amount under these two transactions are 3.5 million to be paid to me, and 1.5 million to be paid by me to Oil bank. But this has quite large capital implications for oil bank. Oil bank needs to account for 3.5 million that it has to pay me. And it also has to allocate capital for the 1.5 million that I have to pay it. This is pretty expensive for the bank (and for me as well, because i need to deposit margins). But if this is a single agreement, all the exposure can be netted off to a single payment of 2 million that Oil bank pays me.

As in all examples, a) this is a gross simplification and b) This is only a single building block.

The ISDA has lots of conditions before you can net exposure across transactions like how I mentioned. However, this is the first step to make sure you can!

Next post, we move to Section 2. Yippee, only 13 more sections to go! (And thats the easy bit). The Schedule is the one that will eat up quite a bit of time, so I will dally less and go through conditions more quickly. You, dear reader, will have to keep up!

Derivatives, ISDA, Tutorial, Uncategorized

Derivative Documentation — Post 2

I would suggest starting with the earlier post in this series if you are new to this and have not had any exposure to Over The Counter (OTC) derivatives.

With that disclaimer done, lets dive into the documentation. The first question to ask is what documentation is needed, and why? Then we move into how this documentation is executed

Derivative transactions are contracts

Any OTC derivative transaction is a contract signed between 2 counter-parties. Thus, it may seem intuitive that there would be a simple contractual agreement that defines the terms to sign, and that would be that. However, that does not work for a few reasons.

Why are Derivative documents complicated?

When a bank lends a company money via a term loan, this is usually a one-time thing. In that instance, a single loan agreement that contains all the terms and conditions of a loan would work. This is because a loan is sanctioned only once, and any drawdown of the loan by the borrower is done under the terms of the signed agreement. There is no further documentation that is needed.

However, derivative transactions are rarely one-time transactions. In most cases, the client would enter into multiple transactions. In the exporter example given in the previous post, an exporter would enter into a derivative transaction to hedge every contract to supply his guar gum that he exports. This would mean multiple transactions, with different values, different maturity terms, and even different hedging strategies.

Each transaction would be a separate contract, and would have to have its terms and conditions clearly defined. We can do this with each and every derivative transaction that is entered into, but this is not really an optimal solution.

A better way to solve this is to have a single agreement that identifies common terms and conditions. This would only be signed once. Then, when individual transactions are executed, the two CP’s would sign the commercial terms of that particular transaction.


The figure above shows a simplified version of the documentation that would then be executed.

  1. Master Agreement — signed before transactions are done
  2. Individual transaction confirmation — signed at the time of individual transactions
  3. Security documentation (collateral documentation, charge on assets, etc) — could be signed along with the Master Agreement, and updated in case of need at the time of specific transactions.

So what does each agreement contain?

Master Agreement

  • Identifies the two counterparties
  • Specifies types of transactions covered under the agreement
  • Identifies jurisdiction as well are relevant laws governing the agreement
  • Specifies triggers that terminate one or more transactions under the agreement
  • and many more things

Transaction Confirmation

  • Specifies the type of transactions
  • Identifies trigger events that will trigger cash flows
  • any transaction specific terms and conditions

Security Documentation

  • Can define types of acceptable collateral
  • Charge on specific assets

ISDA Master Agreement

Master Agreements can be of any form or format. However, in the interests of standardisation, a bunch of large counter-parties banded together to make a common document format that was like vanilla. Everyone was ok with it, but no one loves it!

This group of people called themselves the International Swaps and Dealers Association (ISDA). They made an excellent master agreement form many decades ago, which has been mildly modified a few times. Unfortunately, the entire site is behind a paywall which is a mile high which deters the weak willed (or at least the proletariat) from accessing it. Luckily for us, there is a way around. And it is the US regulator, the SEC that provides us a way forward for our next set of tutorial posts. Here, is an SEC link to an actual executed ISDA between Bank of America, and LKQ corporation. We will use this to understand how the agreement works in the next installment of this tutorial.

Derivatives, Tutorial, Uncategorized

Derivative documentation — An Introduction to ISDA Master Agreements

First up, a disclaimer. This is the first in what I hope will be a long series of pieces about derivative documentation.

Target Audience: The finance professional who needs to also know documentation

What you should expect to learn: What is an ISDA agreement? How does it make a difference to you?

What does this post have: The first post explains what an OTC derivative is. the next post will provide an overview of the documentation.

Introduction to OTC Derivatives:

The first questions to ask are these. What are OTC derivatives? How are they different from the calls and puts and the futures contracts that i trade on the NYSE or the NSE or a thousand other exchanges?

OTC Derivatives are derivative contracts that are not traded on an exchange. They are individual contracts between two counterparties, and do not have an exchange mediating the transaction.

This means that OTC derivatives are not freely tradeable. This allows specific modifications to be made between parties to reflect either transaction complexity or client complexity.

An Example

Transaction complexity: Lets say you export guar gum to a shale oil company in the USA. They pay you in USD. Your raw material is guar which is grown in Maharashtra in India. The catch is that the client pays you USD only after 3 months. You would like to hedge your currency risk by locking in your USD receipts with an option contract. In this case you would buy a put option @ INR 65.65. So this way when you get your USD from your client, you sell your dollars at 65.65 to the other client. If the price is more than INR 65.65, since it is a put option, you have the option of simply selling it in the open market (at say 66, which is the market price on that day). If the price is less than 65.65 (say 63), you have saved yourself INR 1.65 for every dollar (that is a good 2% extra margin, in an industry where usual margins are 6%.

The problem is that options are not cheap. Your 1.65 rupee profit on the option contract is probably 70% eaten up by the premium you pay to the CP.

So you can reduce your cost by selling a 63.5 Put option. Yes, you would be in trouble if the rupee strengthened  by more than 1.75 rupees. But your cost of the option drops significantly.

Now, this sort of structure is very hard to do with exchanges. Especially if it is not a standard value or a straightforward time period. So you need to search for a CP who is willing to match your exact value of your export proceeds, and also the time frame in which the client will pay you.

In steps in a market maker (usually a bank). They will provide you a customized rate, as well as a customised product, and you can pick and choose the exact type of product you want to buy.

This is an over the counter product, rather than an exchange traded one, hence the term OTC derivative.

The other side of this being a customized product is that it is not liquid. You will find it hard to sell this option strategy once you buy it. So, if your export order gets modified, then you have to jump through hoops if you want the hedge to keep up with the updated contract terms.

Counterparty Differences

Also, the bank that you do this deal with is going to evaluate the client types differently. If you are a simple exporter who has receipt of USD and you only want to hedge this risk, the bank will have a totally different set of checks than if you were a macro trader who was going to take a wager on the creditworthiness of Greek sovereign debt.

However some documents are consistent across all types of counterparties. We will start with the first one in the next episode of this series.


Trust in Banking and Business

About 10 years ago, I got into a bit of a spat with one of my professors. He and I had a rather fundamental difference of opinion on the role of trust in business transactions, especially microfinance.

My argument is that in a peer pressure situation, when collateral is not available, use the peer pressure and trust the borrower to pay you back…else the whole group that acts as a guarantor gets penalised. My dear professor of course begged to disagree, and said that only an “NGO like Grameen Bank can do this”.

That was a rather strange argument. As long as the institution is making money…who cares what its called? Grameen Bank was profitable, and made money based on this business model. Sure, it had its own set of issues, and repeating this business model was a challenge. But this does not imply that trusting your borrower is bad!

And this is a point that can be made in business. Trusting business partners and suppliers makes hard nosed business sense. Trusting your supplier ensures that you do not have to duplicate processes to check defects. Trusting your customers means that you spend less on fraud detection. Trusting your employees means HR is seen less as a gatekeeper, and more as an enabler.

It is now 10 years later, and my business school prof has long since retired. I have had some years of working with actual customers and suppliers to consider my views, and my professors arguments more carefully. And quite co-incidentally, news broke about a fortnight ago about a 2 billion dollar fraud at Punjab National Bank. This fraud appears to have benefited a customer of the bank, and appears to have been enabled by a few (or many) insiders at the bank.

A bank operates a great deal on the basis of documents and process. A banker trusts documents. He (Unfortunately almost always it is a he) tends to rely on invoices and past performance certificates from auditors. And for good reason, becuase banks are custodians of public money.  Their every decisions is made with the knowledge that it will be scrutinised by a hawk eyed group of supervisors, auditors and regulators, in that order.

However, none of this seemed to have worked at Punjab National Bank. It appears that the problem was inadequate controls and checks to verify if the client was honest, and connivance from insiders. In the aftermath, bankers and regulators are busy making new checks and new processes so that this sort of failure does not happen again.

They all operate under the assumption that the bank is a honeypot, that attracts the theiving employee, who will always put his hands in the pot to grab as much honey as he can. ear of being stung by bees is the only way to keep the banking employee on the straight and narrow.

This in my view, misses the wood for the trees. The fraud was certainly committed by a crooked employee, and definitely all checks and balances failed. But the solution is not necessarily to tie employeee hands to the wheel of process. Instead, it may make sense to ‘trust, but verify’, in the words of an ex president of the United States of America.

How would this work in practice? In the banking scenario, we have a concept of a maker and then a checker to avoid errors. But this is not really a fraud prevention measure (or should not be). It would be prudent to have a periodic review of outcomes as a verification process. In the PNB case, the outcomes to be measured were whether the debits and credits that actually hit the banks accounts were actually reflected in their core banking systems. A daily check of this would have caught the fraud by the end of the second day, instead of after 8 years.

How would a regulator act? In India (and in most developed economies), regulators have a relatively hands off approach until the proverbial horse manure hits the rotating air circulating device. Yes, Basel guidelines have established a concept of operational risk based capital guidelines. But in my opinion, this answer doesnt go far enough. A recent action by the US Fed in Wells Fargo may give us a possible way forward though.

Going into the web of malfeasance at Wells Fargo would require its own blog post, but we will stick with a summary. Management made a bunch of mistakes in the drive to push profits. These errors led to customers being bilked. The regulator initially fined the bank, which the bank duly paid. Outrage was lobbed at the bank management, and a couple of senior managers duly fell on their (metaphorical) sword. But I doubt things would have changed…until the last response came from the Fed. It barred Wells Fargo from growing! This is a great example of outcome based management. Management did a bunch of cutting corners in the quest for growth, profits and big bonuses. Explicitly telling banks (and shareholders), that if you cut corners to get growth, you lose whatever you aimed for. That’s outcome based policing for you!

Outcome based verification is not an easy job. It is also very likely that it will not be a silver bullet that will solve all your trust problems. It is likely to be slow, and won’t satisfy the urge for quick justice to be meted out. But when it works, it changes incentives for bad behavior. And by monitoring the outcome, you can make it possible to trust your employees in their day job.