Global Policy Watch: Accident Ho Gaya
Insights on global policy issues relevant to India
— RSJ
I must admit there are times when I have made a big deal about writing this newsletter. Not about the content, mind you. I’m not that vain yet. But the regularity of it all. Getting about 4000 words out between the two of us every week isn’t trivial stuff. But then there are weeks I ask myself if it is really a big deal. I mean, there are weeks when there’s so much happening in policy, politics and macro spheres that things just write themselves. I have been in what could be called writing self-help groups where people bemoan their writer’s blocks and the soul-crushing experience of staring at a blank word document with the cursor blinking. To them, I have two pieces of advice. Switch to writing on public policy. And don’t bother much about quality (speaking for me here, not what Pranay produces).
Voila! You get something like 50% of this newsletter.
Anyway, coming back to this week. Sometime midweek, I thought it might be a good idea to write about the state of opposition in India in the light of Rahul Gandhi’s Bharat Jodo Yatra and his media engagements in Oxford. There’s always space for the opposition in India despite the brutal electoral majority of the party in power, as we have seen in the past. This was evident during the yatra. What is also evident now is that there’s a complete lack of understanding on the part of Rahul Gandhi on issues that can animate the electorate. So, he can only hope for the party in power to hit self-destruct mode to score an electoral victory. What’s worse is he has terrible ideas of his own and a tin ear for good advice. His acolytes defend him saying he’s sincere. I can only say when you combine sincerity with bad ideas, you get demonetisation and instant lockdowns. Back to the point. As I was thinking of writing about this, news came in of Manish Sisodia, the deputy CM of Delhi, being taken into custody by ED for what’s being called the liquor scam. Liquor policy in various Indian states is a gift that keeps giving. We love talking about it. What I also thought was admirable is the agile way the ED functions these days. Like some start-up in Koramangala. It is always hustling.
These were the ideas I was toying with till a bank with a balance sheet size of US$ 200 billion (a tad smaller than HDFC Bank) collapsed in the US. And, so, served on a platter was another possible post on what could go wrong in the global economy. I’m afraid Rahul Gandhi, Sisodia, and liquor will have to wait for another day. I would like to discuss the aptly named Silicon Valley Bank (SVB) that has gone from boom to bust in less than two years.
Here’s what has happened since Thursday. WSJ reports:
“On Wednesday SVB said it had sold a large chunk of its securities, worth $21 billion at the time of sale, at a loss of about $1.8 billion after tax. The bank’s aim was to help it reset its interest earnings at today’s higher yields, and provide it with the balance-sheet flexibility to meet potential outflows and still fund new lending. It also set out to raise about $2.25 billion in capital.
Following that announcement on Wednesday evening, things seemed to get even worse for the bank. The share-sale announcement led the stock to crater in price, making it harder to raise capital and leading the bank to scuttle its share-sale plans, The Wall Street Journal has reported. And venture-capital firms reportedly began advising their portfolio companies to withdraw deposits from SVB.
On Thursday, customers tried to withdraw $42 billion of deposits—about a quarter of the bank’s total—according to a filing by California regulators. It ran out of cash.”
Looks like a good old run on the bank. The regulators had to step in. Again from WSJ:
“The Federal Deposit Insurance Corp. said it has taken control of the bank via a new entity it created called the Deposit Insurance National Bank of Santa Clara. All of the bank’s deposits have been transferred to the new bank, the regulator said.
Insured depositors will have access to their funds by Monday morning, the FDIC said. Depositors with funds exceeding insurance caps will get receivership certificates for their uninsured balances, meaning businesses with big deposits stuck at the bank are unlikely to get their money out soon.”
If you’ve been reading me over the past month, I have made three points. One, it is foolish to assume that the Fed will pause on rate hikes anytime soon. Inflation isn’t transitory in the US. And this was made clear this week when the Fed Chair, in his response to the questions from the Senate Committee on Banking, said that interest rate hikes are “likely to be higher than previously anticipated” and that because of it, the labour market is also likely to weaken in the near term. Like I have said before, it is best that markets, banks and companies plan for scenarios where the rate goes up to 7 per cent to stress test their models. This holds for India too.
Two, it is inevitable that a sharp rise like what we have seen in the past nine months will mean there will be ‘accidents’. We don’t know yet what their nature will be, but they will mostly emanate from private markets and the ‘new’ asset groups (like crypto, VC funded business models) where bubbles have built up.
Three, whatever accidents happen may not lead to a contagion. They will mostly singe private markets or those holding these new speculative assets.
In a way, what has happened with SVB this week bears my thesis out.
So, what happened? And what does this mean?
During the pandemic, the US treasury pumped trillions of dollars to keep the economy afloat. But because of the pandemic, there was nothing to spend this money on. So, the money found its way into banks as deposits. The total deposits in US banks went up by $ 5.5 trillion, of which only about 15 per cent could be used to lend because of weak demand. So, what could Banks do? Well, like prudent entities, they parked this surplus in securities or kept it as cash. Now, when Banks buy securities, they are asked to take a call upfront on whether they plan to hold them to maturity. This decision then labels the securities as either held-to-maturity (HTM) or available-for-sale (AFS). Simply put, with an HTM security, the bank is declaring that it will hold, say, a US government 10-year bond till they mature (that is for ten years), while if it categorises another bond as AFS, it means it can sell them anytime in between.
Now, from a regulatory perspective, this plays out in different ways for a bank. An AFS security gives banks flexibility to sell a security if the world changes around them, while an HTM security allows banks to weather a fall in value because they aren’t marked to market (M2M). Therefore, they will remain on their balance sheet at amortised costs regardless. Banks don’t have to crystallise their losses on HTM portfolio because the expectation is that on the date of maturity, they will receive the full redemption value. However, if the bank sells anything out of its HTM portfolio, it has to reclassify the entire portfolio as AFS. In 2020, US banks had about 74 per cent of their portfolio in AFS securities. As things opened up after the pandemic and inflation started taking root, interest expectations rose.
A small explainer will help the lay reader here. Others can skip ahead.
Bond prices are inversely related to interest rates. Suppose you bought a 1-year government bond of Rs. 100 with a coupon rate of 5 per cent. This means at the end of the tenure, the bond will fetch you Rs. 105 regardless of the bond's underlying market price. Assume the prevailing repo rate (benchmark interest rate) was 4 per cent. What this meant was you had an incentive to purchase this coupon that would give a little extra over the prevailing benchmark rate, and at the end of the year, you could redeem the bond and receive Rs.105. Because many people buy and sell government bonds, there is a market for them where they get traded. Now consider a scenario where in the middle of the year, the benchmark rate was increased to 10 per cent by the RBI. What if, for any reason, you want to sell your bond during this time? Nobody will buy your bond for Rs. 100 which is the price you bought it at. Why? Because it will fetch them only Rs. 105 (5 per cent return), while if they bought a fresh 1-year bond priced at Rs.100, they could get Rs. 110 (10 per cent return). But you’re desperate for money, so the best you can do is to offer your bond at Rs. 95.50 or so. Because then whoever buys it will get Rs.105 at the end of the year and will make an equivalent 10 per cent return. So, you take a loss of Rs.4.50 (or about 4.5 per cent) by pricing your Rs.100 bond at Rs.95.50. At an aggregate level, this is what happens in the market as the interest rates go up. The price of the bond falls. And conversely, if the rates fall, the bond price goes up. So, a Bank holding securities sits on a mark-to-market loss or profit depending on whether interest rates go up or down.
As interest rates rose, the US banks that were sitting pretty with mark-to-market gains on their AFS portfolio started having losses emerge. Now you could keep taking mark-to-market hits every quarter, or you can reclassify the AFS to HTM, take a one-time loss upfront and move on. Because once you make it HTM, remember, you don’t have to account them as marked-to-market. And that’s what has been happening for most of 2022. Banks reclassified their AFS, and their share fell from 74 per cent in securities portfolio to a little less than 50 per cent.
The Case of Silicon Valley Bank
Silicon Valley Bank (SVB) had an interesting time during the pandemic. It has, over the years, positioned itself as the bank for the valley ecosystem of startups, VCs and angels. The boom in VC funding since 2019 meant its deposit balance tripled to about $200 billion by March 2022. What’s more, a large part of it was demand deposits which don’t bear interest. As we have learnt now, its cost of deposits or the interest it paid its depositors on average was about 1.2 per cent. What does a bank do if it has so much money flowing in? It has to put the money to work. But SVB’s problem was there weren’t too many loan takers in its customer base since it catered to the valley ecosystem flush with funds. The real and safe option it had was to put money in securities. That’s exactly what it did. Around 12 months back, it had about $100 billion in HTM and another $25 billion in AFS through various securities like the US treasury bonds and mortgage-backed securities.
That’s when the Fed's action on interest rates started. And it was rapid. This meant the price of those securities, especially the mortgage-backed lot, fell quickly. By September 2022, the unrealised losses on the portfolio had gone up to $16 billion, while the total common equity capital was $11.8 billion. Technically, the bank was insolvent. But these losses on the HTM portfolio don’t have to be recorded on the bank's books. What a Bank would do is wait for bonds to redeem, hope for interest rate hikes to pause, keep their depositors with them and see through this tough time. It was doing exactly that except for one more wrinkle. Its deposit portfolio was skewed to the fortunes of the valley. As funding dried up, the valley companies burned through their bank balances. This meant they were constantly withdrawing money. From $200 billion in March 2022, the deposits fell to $165 billion by last month. There was a serious danger of the bank not having liquidity to give depositors their money unless they restructured their balance sheet.
Endgame
This is what SVB set out to do last week. They sold $21 billion of AFS at a loss of $1.8 billion to have cash available to them. And to recover this loss, they planned to raise capital by issuing fresh equity. Tough to explain but still doable. As luck would have it, the same day, a crypto bank, Silvergate Capital, announced it was insolvent for pretty much the same reason, except that SVB had real cash deposits. The depositors lost their nerve. Large VCs called for their portfolio companies to take money out of SVB. In the good old days, this would have meant going to branches and asking for your money. It would take time. Today, it is just a few clicks on the mobile app, which is aimed to give you a frictionless experience. That lack of friction meant customers tried to withdraw $42 billion - about a quarter of its total deposits - in a single day. Most of the depositors at SVB had balances of more than $250,000, which is the threshold for deposits to be insured. This meant the demand for withdrawals was real. The bank ran out of cash. The CEO came to assure depositors that they needed to have patience and the bank was liquid. I might have told you before - the moment a bank has to tell its depositors that they are liquid, they won’t stay liquid any more.
What could the bank have done? Its HTM book could be sold. But selling a single bond there would have meant marking the whole portfolio to market. That would have been a huge loss, and the bank didn’t have the capital to absorb that. It could have borrowed funds, but that was coming at closer to 5 per cent, which was expensive. The only real thing it could have done was probably two years back. It could have diversified its depositors base, figured that its asset portfolio was too skewed to interest rate risk and reduced its balance sheet size. There are regulations on the nature of your deposits through what’s called Liquidity Coverage Ratio guidelines, but SVB was deemed too small for it. It is a lesson in policy making that what’s small or big is related to the broader macro context and cannot be fixed forever. This is what I meant when I said the scenarios that could unfold because of a rapid rise in rates are unknown to us. I mean, look at the percentage of unrealised losses to total equity among US banks. There are some vulnerable banks there. Is someone asking the same question in India? As I mentioned a couple of editions ago, it will be good for India to plan for a scenario where the Fed hikes the rate all the way to 7 per cent.
The Fed cannot anticipate the unintended consequences of its battle to tame inflation. At some level, I guess it doesn’t care as much. If there are weaker players who get sick on this ride up, so be it. The accidents are waiting to happen. We will be plain lucky if the contagion is contained to private markets because we believe the larger public markets and banks are better regulated. But you never know.
Matsyanyaaya: The Indo-American Dance
Big fish eating small fish = Foreign Policy in action
— Pranay Kotasthane
There’s been a flurry of Indo-American diplomatic activity in recent weeks. The two National Security Advisors (NSA) held the inaugural meeting of the initiative on Critical and Emerging Technology (iCET) on Jan 31. Then there was the mid-February phone call between the Indian Prime Minister and the US President. In early March, the four foreign ministers of the Quad made it a point to turn up at a Raisina Dialogue panel. Most recently, the US Commerce Secretary Gina Raimondo was in Delhi, where she held meetings with the Finance, Education, External Affairs, Defence, Commerce, Electronics & IT ministers, and the NSA. The two commerce ministries also signed an MoU on Semiconductor Supply Chain and Innovation Partnership.
These are all significant shifts. Perhaps we can forgive Mr Singh’s Holi dance performance, where he appeared as anatopistic as Bharat Bhushan would have felt in a Karan Johar movie.
In edition #165, I analysed the India-US relationship using a tri-axis framework: state-to-state relations, state-to-people relations, and people-to-people relations. The people-to-people relationship never had a problem, to begin with, and state-to-state relations have never been better. Yet, surveys suggest that Indians continue to be circumspect of successive American governments.
Perhaps for a good reason. One can’t deny the US stance in the 1971 war, its continued support to Pakistan despite the latter’s anti-Indian projects, and its role in the multilateral export control regimes that held back India’s space and nuclear programmes.
These three reasons became the foundations on which the edifice of anti-Westernophobia was constructed. Like other ideologies, instances that reaffirmed the ideology got amplified and internalised, while instances that didn’t fit into the dominant narrative were discarded.
For example, consider an oft-repeated argument: a closer partnership with the US would imply that “India will be dragged into its wars.” The logic is as follows. The US, as a superpower, keeps overextending itself as it’s an integral part of its strategic doctrine to tackle the adversary before the threat reaches its shores. This means that partners have no choice but to fight these expensive and sometimes irrelevant wars. If they dare to strike a discordant tone, the US will use its immense power to punish them.
The problem with this story is that India’s own experience points out otherwise. India did face this question after the US misadventure in Iraq. By then, the relations were on an upswing. The Bush government wanted to hand over the post-war transition to a “coalition of the willing”, and India featured prominently in those plans. The Indian government considered this request at the highest levels and eventually chose not to intervene. Despite this refusal, the India-US relationship didn’t face a major bump. The two countries announced an initiative called the Next Steps in Strategic Partnership (NSSP) six months later. By 2005, the two countries had enough confidence to sign the monumental civil nuclear deal. So much for the concern over dragging India into American wars.
The civil nuclear deal provides another counter-point. A recent book by veteran journalist Seema Sirohi Friends with Benefits: The India-US Story, meticulously details the events and personalities involved from the American side. What struck me most was the efforts made by the Bush administration to align cognitive maps of domestic and international opponents, that too within a decade after the post-Pokhran economic sanctions had created a new low in the relationship. On the issue of getting the NSG waiver past six opposing countries, Sirohi writes:
“The Americans used every weapon in their diplomatic arsenal to persuade countries. At one point Mulford suggested sending a warning to Austria that Washington was ready to tell India to cancel all Austrian Airline flights. The British and the French, who were helping the US, would corner recalcitrant European ministers at every opportunity in different locations—including in opera houses and trains—to argue for the NSG waiver. In the end, all forty-five were persuaded. Many saw India as a responsible nuclear power with the potential to play an important role in world affairs.”
Now, none of this was out of benevolence. In fact, to use a construct applicable to human relations in state-to-state relations would be grossly inappropriate. To say that the US is hypocritical, unreliable, or benevolent makes no sense in an arena where national interests are supreme. Anthropomorphism in international relations is as sound as counting goals in a cricket match.
Rather, what this instance illustrates is the benefit of a closer India-US relationship. Having the world’s number one power on the same side can open doors that strategic autonomy can’t. To be sure, there will be costs. But, given that the interests of India and the US have never before converged as they have today, India’s mileage out of the relationship would likely be much higher. After all, Taiwan, Japan, South Korea, and even China gained power in no short measure due to the trade, technology, and human flows from the US. This is India’s opportunity to amass national power quickly.
PolicyWTF: Number Mirages
This section looks at egregious public policies. Policies that make you go: WTF, Did that really happen?
— Pranay Kotasthane
Don’t let numbers fool you. This theme has spawned a new genre of books, starting with Darrell Huff’s 1954 classic How to Lie with Statistics. And yet, governments continue to insult our intelligence by using the same tropes repeatedly. What’s worse is that they often succeed.
So in this week’s policyWTF section, let’s look at some favourite number-chicanery tools used by Indian governments.
One: present a current figure as a multiple of its value from almost a decade ago.
Consider the news item from this week. “According to the National Statistical Office (NSO), the estimated annual per capita income at current prices for 2022-23 has doubled since the Narendra Modi-led NDA came to power in 2014-15”. Presented this way, the rise in national income sounds really, really impressive.
Now pause for a moment. Using the shorthand rule of 72 suggests that a figure increasing at roughly nine per cent annually doubles in eight years. Instead of putting out the 9 per cent annual growth, the government showed you national income as a multiple of its value eight years ago.
Two, present numbers at their nominal values in place of inflation-adjusted values.
Another common strategy to make small increases seem bigger is to compare numbers at current prices rather than constant prices. The national per capita income story from this week uses this second trick as well. Adjusting for inflation, the incomes have only risen by 35 per cent in the last eight years, which translates to a poor 3.8 per cent annually.
Instead of pointing out this abysmally low number, most news analyses went on a tangent to explain how the national per capita number is a mean value and doesn’t account for the variance in incomes (duh!). Soon it became a debate over growth versus income inequality. The government went scot-free.
Three, use a data point and its converse, both, to claim success.
Thanks to RSJ for pointing this one out. Take a recent example. Demand collapsed at the height of the pandemic, and so did imports. That led to India showing a current account surplus for a few quarters, a rare occurrence. The government celebrated this “achievement”, downplaying the worrying cause of the surplus. When the current deficit was back when the pandemic ended, that too was celebrated as an indicator of strong domestic demand and bounce back!
Four, spew out decontextualised metrics.
Using absolute numbers instead of per capita numbers is an old trick in the government. Most numbers related to the government are big. If I were to tell you that the Corporate Social Responsibility (CSR) Law (aka Mandatory Philanthropy) generated Rs 21,000 crores in a year, you might be impressed. But only if I were to tell you that this is less than half the amount the Union government spends on MGNREGS alone would you ask: are the compliance costs worth the benefits of the CSR law?
Five, compare budget estimates of the next financial year with budget estimates of the current financial year.
When budget documents are presented in the legislature, governments have revised estimates of the current year with them. An old trick is to compare the allocation of the next year with the older budget estimate in order to paint a better picture. For instance, the press release for the defence budget this year reads, “Defence gets Rs 5.94 lakh crore in Budget 2023-24, a jump of 13% over previous year”.
What’s left unsaid is that the increase over the revised estimates is a mere 1 per cent, that too in nominal terms. Accounting for inflation, the government will spend much less next year than it spent in the current fiscal year. The headline is perfectly accurate and perfectly misleading, both.
This list is nowhere near exhaustive yet. Are there other tricks you recollect?
HomeWork
Reading and listening recommendations on public policy matters
[Podcast] A Puliyabaazi on India-US relations with Seema Sirohi
[Book] A classic - FA Hayek’s The Fatal Conceit
[Podcast] This Ideas of India episode is a treat for anyone interested in urban governance and planning.
[Article] Shekhar Gupta’s take on India-US-Russia relations.
Share this post