In your opinion, what does this blog need more of?

Saturday, October 12, 2013

On Bank of Nova Scotia's new Indian gold deposit scheme, part two: gold strikes back

As I said in my previous post, an interesting news item hit the wire over the past few days. Here's a link for you to read first:

Reuters - India's top bullion bank to work with jewellers to tease out gold hoards.

So basically, BNS is offering a gold term deposit scheme that will pay out 2.5%-3% interest, in gold, and send that gold to the jewellers so they can melt it down to make and sell jewelry.

Because it's Indian gold to begin with, this eases the need for imports, so India's CAD drops.

Again, here's a quote from the Reuters article:
Scotiabank's plan aims to enlist jewellers to collect the gold, which could make it more accessible in a country where many people use family-run jewellers for generations and banks are few and far between in rural areas.

Other attractions of the scheme could include more flexibility in the duration of deposits plus tax-free interest.

Soni said he had proposed a lock-in period for the deposits with Scotiabank of two to seven years, compared with the three to five years in a similar scheme run by state-owned State Bank of India (SBI).

Indians often use gold as a ready source of liquidity in times of need, making longer-term deposits less attractive.

SBI offers interest rates of 0.75 percent to 1 percent on gold deposits, depending on the time period.
In my past post I tried to puzzle out exactly how much Indian gold this would bring into the scrap market. Because that's essentially what this is: BNS takes your gold via your local village jeweller, gives you an IOU and 2.5%-3% interest over a 2-7 year term, and then melts down the gold and sells it back to jewellers.

If it actually works, then the scheme helps the jewellers in increasing the amount of local-origin scrap gold available for resale. So they don't have to worry about the stupid import restrictions.

And it helps the Indian government, because local-origin gold will then replace imported gold, so their (dangerous and unsustainable) CAD goes down.

But maybe it doesn't, and here's maybe why.

The gold deposited is melted down and sold. Paper replaces it. The interest is paid in gold. The principal gold eventually has to be paid back at the end of the term - at least in theory.

So Rajesh is essentially lending BNS 100 grams of gold. He has to be paid 3 grams of gold per year over seven years as interest. BNS sells the 100 grams to a jeweller.

If Rajesh decides, after seven years, not to roll over his principal into a new deposit, then after seven years BNS has to buy it back from somewhere to close out Rajesh's account.

Obviously that means BNS is going to have to hedge against future gold price increases. If gold continues to go down (in rupee terms) from the date of deposit to the date of maturity, then BNS is making a profit on the deal; but if gold goes up, BNS has to hedge out that loss. It gets hedged, period.

I'm no economist so I don't know what that does to the price structure for gold futures. You explain it to me. I'm assuming that enough hedging against future gains drives gold higher over time, since it's been asserted that gold miners' past hedging against future losses was a downward price pressure on gold. But you tell me if I'm wrong.

Another thing that springs to mind with this is that after Rajesh's 7-year term is up, he's already been given 21 grams of gold in interest, plus he has the right to ask for his 100 grams principal back. But meanwhile, those 100 grams of principal have been melted down and sold to somebody else.

So after 7 years, BNS has to buy 100 grams of gold from someone to pay Rajesh back.

This is fine if BNS is successful in encouraging more gold to come in per year than is leaving via redemptions and interest. But per my last post, I don't see massive participation in a gold deposit scheme that pays 30-40% of the Indian 10-year bond and exposes the average Indian's prized asset to de-allocation, encumbrance, loss of liquidity of an asset, and probably a whole pile of political risk.  Plus jewelry can't participate (I assume) because its sizeable heirloom and craftsmanship premia are destroyed.

But if this scheme works at luring in enough Indian gold, then it isn't destroying Indian physical gold demand; it's increasing it. Gold interest is being paid out, more gold jewelry can be sold, the gold term structure is steepened, and now gold "the worthless asset that just sits there" is also becoming an asset that pays out interest.

Frankly I don't see how BNS can look at global supply/demand trends and think that this scheme will be profitable long-term in any way.

Another possible problem I see with this is that, when the next Indian financial crisis comes, instead of it causing a rush to sell scrap, it'll cause a rush in withdrawal of gold deposits. BNS will have to buy that gold on the market, and thus instead of a spike down in gold prices from a pop in scrap, the world market might see a spike up. Which I guess is what we all want to see gold do, but which it hasn't actually done in decades.

And, in case of domestic crisis, those poor Indian peasants who thought they were being all clever and modern by setting up gold deposit accounts are going to realize how stupid they were. Is the deposit locked-in for the term? Then it'll really suck for you when you have a really shitty harvest (or a cyclone washes away your village, ahem) and you've naively traded in your physical gold for a piece of paper in a box somewhere in Delhi. Sucks to be you, this year's 3% payment comes in December, have fun starving til then.

Another problem might be that India could change the rules. What if BNS runs to the Indian government, say during a time of crisis that causes a gold deposit run, and whines that they can't possibly find 2000 tons of gold to pay back to all the depositors who are clamouring for their gold back? Well, the government's already done douchebag things like retroactively levying 5 years of back taxes on Nokia; why do you think they won't ever change the gold term deposit rules retroactively so that these accounts only need to be paid out in rupees?

In a crisis, that's exactly the time when the Indian depositor's preference for rupees over gold is at its lowest.

That last problem goes to the heart of the "political/currency risk" problem; is any Indian seriously going to value the political risk premium alone at 3%? Seriously?

But if this sort of gold-run-causes-deposit-default scenario ever were to happen, you can bet it would be a tremendous setback for India's progress toward a modern banking system. There would be yet more justification for the average Indian peasant to hold all his savings in physical gold, and to avoid any participation in the Indian banking system.

Meanwhile, if you did get a massive wave of participation in this gold deposit scheme, remember that all this gold gets melted down and sold to jewellers. That means if 2000 tons of gold rushes into these deposits, the net world supply of gold increases. But does BNS get to mark their gold obligation to market? I'd like to know how that's going to work.

And if the price of gold drops after the initial deposit, your payout (in gold) has now dropped even lower than the 2.5%-3% initially offered. All of a sudden the average Indian sees this as even less of a deal than it originally was.

And if the net world supply of gold increases, does that simply make EMs buy more gold? It'll be cheaper, after all. But with dwindling world supply, actually with zero mined supply if the gold price crash gets strong enough, what does that do with futures?

Looking at the above, it seems that BNS's gold deposit scheme is little more than borrowing present gold against future gold. And India is a big enough country, with enough of the world's physical gold, that any real participation in this scheme will really tilt the fuck out of gold futures.

Sure, a big rush in by all Indians would crater the present-day price; imagine India putting 2000 tons of gold into these gold-deposit schemes. But that would still mean BNS would have to pay out 60 tons of gold a year in interest, plus those 2000 tons of principal would be an evident liability unless they could keep the rollover rate really high.

Meanwhile China, and EM central banks, keep buying gold. While mineable deposits are dwindling.

Which, I guess, is why BNS can't possibly offer a sufficient interest rate to make it worth anyone's while.

Given the above, you'd have to expect that BNS, if a rational actor in all this, is going to try and find the sweet spot in participation that stabilizes gold prices short-term (i.e. not too much of an inrush of physical) and long-term (i.e. no net effect on long futures). I would assume it doesn't want to cause chaos either way in gold prices, or else the volatility premium on its hedge goes sky-high.

Maybe I don't understand this enough, and I probably would have to be an investment banker to puzzle this out sufficiently. But after thinking over these two posts, I don't see why the BNS Indian gold deposit scheme should have any net effect on gold prices.

1 comment:

  1. Maybe it's like the Gold Guarantee was in Singapore. BNS (which the French call the "Bank of Silly Notions") will pay out one year's interest and then disappear.