Let us say there is a great business you would like to set up that involves a return of 20% every year after expenses and taxes. What if I gave you two choices:
1) You put in Rs. 1,000 as your investment, in setting up a business that returns Rs. 200 per year.
2) You put in Rs. 200 and borrow Rs. 800. The business still generates Rs. 200 per year. You pay Rs. 80 as interest on the loan, and the remaining, or Rs. 120 is yours.
In the first case the return you get is 20%. Your capital is at stake so if the business slows down to a return of 5%, you'll only get Rs. 50 per year.
In the second case, you make a 60% return — Rs. 120, on Rs. 200 of your capital. Awesome? But if the business drops to a 5% return, you will end up getting Rs. 50, and you now have to pay interest of Rs. 80 — for which you have to shell out more money from your pocket. Effectively that whittles down your capital to Rs. 170 (Rs. 200 minus the 30 as excess of interest over profit)
In the second case, the return was amplified by the loan you took, and the losses were amplified as well. This is the classic concept of leverage, where you increase returns by using money you didn't have.
Leverage is a two sided beast. We all like the upside — amplified returns — but we don't like the impact of huge losses. Leverage is unavoidable today, and if you have money in a bank, even if you don't have a loan, you're impacted.
Let's take the case of an economy with a single bank.
I put in Rs. 100 into that bank. The bank takes that Rs. 100 and keeps Rs. 10 as a Reserve, because of a reserve requirement of 10%. It then lends Rs. 90 out to an Amit Varma, who expects to take that money and double it with his business. But until he does so, he puts the money in the bank (remember, there's just one bank).
So my Rs. 100 has become Rs. 190 already. I have Rs. 100 and Amit has Rs. 90 in the bank. The bank owes me Rs. 100 and Amit owes the bank Rs. 90. So the amount of "money" that's out there is Rs. 190.
The bank, on seeing Amit's money, keeps Rs. 9 as reserve and lends Rs. 81 to that young Aadisht wants Rs. 81 to make his life more interesting in the rural Tamil Nadu setting he is in. The total money increases to Rs. 271.
This can go all the way till the total amount of money people have and owe will be Rs. 1000.
This kind of system is very good in that Aadisht, Amit and so many other people could get money, and more money than the Rs. 100 I had. The bank makes money from the interest that the borrowers pay, offset their cost of sending spam SMS selling me insurance, and the rest is their profit.
You might think the banks have it good - after all, they never put any money in? That's not really true, as sound banking requires them to put SOME money down. How much? They use another ratio, called the "Capital Adequacy Ratio": if this ratio is 10% and the bank has lent Rs. 1,000, then the bank needs to put up Rs. 100 of its own money as capital.
Suddenly this becomes more apparent - for every Rs. 100 the bank has, it can lend out Rs. 1,000. The leverage it has is 10 times.
This is all fine, you think. How does it affect you and me? Because the leverage means there is risk in the system, and that risk is only apparent in a crisis.
Let's just say that Amit's business doesn't do so well, and he can't pay the money back. If the bank says goodbye to the Rs. 90 it lent Amit, and since it still owes the rest of us money, that Rs. 90 will have to be taken out of its capital. The bank now has just Rs. 10 left as capital.
Phew, you think. The bank didn't go bust. Yet, the bank has to now keep to the capital adequacy ratio of 10, which means it can lend only Rs. 100 worth of money! This is the downside of deleveraging — that when you have a problem, the problem magnifies itself on the way down.
What the bank must do now is either to call in all the loans above Rs. 100 — which is usually not possible, because people might not be able to pay. The other option is to raise another Rs. 90 in capital. When a bank is in trouble, no one wants to give it any money. The only way ahead is to generate confidence in the public, by having the regulator or the government say that it will back all the loans up, or to get the banks assets and deposits to be sold to another bank through a merger.
In the 2008 crisis, US banks went through a similar problem, where a large number of (sub-prime) loans weren't getting paid back, and the resulting deleveraging required bank capital — the US Treasury had to feed in capital into some banks (so much that they nearly took over Citigroup) and other sovereign funds chipped in with capital once they had confidence. In India, when Global Trust Bank ended up losing all its capital through bad loans, RBI forced it to merge with OBC, another bank, in order to contain the contagion.
In both the above cases, depositors weren't hit, but in over 100 banks in the US last year, a failure meant that the depositors lost some of their money. To cover such problems, there is deposit insurance — upto $250,000 in a bank account in the US is paid back by the FDIC if the bank "fails". (In India, deposit insurance covers you up to Rs. 100,000)
Let's say I decide to take back my Rs. 100. That reduces the available deposits by Rs. 100 — so the bank now has just Rs. 900 in deposits but it has Rs. 1,000 in loans. It can pay me because it kept that 10% in "reserve". But after me, if everyone tried to take their money back at the same time, the bank wouldn't have the money to pay!
When a large number of depositors try to take money from a bank, it will have a problem — and this is when regulators need to step in and generate confidence. This involves regulators saying everything is fine, which we know, after Enron, Lehman, Bear Stearns and Greece, means that the poo has hit the fan. But without such confidence, bank runs can quickly shut a bank down.
In the US, many sub-prime borrowers took loans for 100% of a house's value, placing the home as collateral. Banks were fine because now if borrowers defaulted the banks could recover the money by selling the house. Except house prices fell, and when borrowers defaulted, banks took the houses and tried to sell — resulting in even lower prices. The spiral of selling then brought down the value of other banks' loans and gave us a crisis of gargantuan proportions.
It is these risks (and there are many more) that call for diligent regulation. For banking, regulators monitor a number of such ratios — the Cash Reserve ratio (CRR) is one, the Capital Adequacy Ratio (CAR) is another, and yet others include non-performing loans (potential defaults), exposure to different sectors (too much to one sector is a problem) and so on. Banks are required to invest nearly a fourth of their deposits in government bonds, so that there is some element of security in the loans they provide.
Still, there continues to be leverage; India is levered about 4x — 5x(depending on which ratio you consider) and the US, nearly 10x. Some European banks have more than 30x leverage, where even a 3% drop in their assets can wipe out their entire capital.
For a highly leveraged system, a crisis is always a crisis of confidence. Just a word that a bank may be in trouble can cause a run on the bank, or for a selling spiral to mark its assets down. Even if the bank is sound, it will fail. That crisis compounds when you realize that no one has any idea of how many other parties will be hurt if one party fails. With banks taking on complex financial instruments and then hiding away from regulation, we've got a system that can go out of control very fast. As an astute saver and responsible unleveraged investor, you may be wiped out just because you choose the wrong bank, or indeed, the wrong geography.
Which is why the whole world waits anxiously as a tiny country votes on whether to pay back its loans. The problem isn't what will happen; it's that the rest of the world will get scared and lose faith. In a world addicted to leverage, we can't have a recession of trust.