Marc Faber is wrong. 2028 or 2038 or later, the rupee is unlikely to ever hit 100

Harsh Gupta Moneycontrol Contributor

In a recent interview, Marc Faber, editor and publisher of "The Gloom, Boom & Doom Report", said "In 1990, the Indian unit was around 12 against the US dollar. In 2008 – 09, it was close to 39 levels and since then, the trend has been down. I am sure the rupee will go over 100 levels. But, will it go over this level in six months or in 10 years is a debatable question. I don’t think it will hit 100/$ in the next six months. But it will hit this level in the next 10 years." I will argue here why Faber is wrong on this specific assertion "the rupee will go over 100... in the next 10 years", not just with respect to the 2028 timeline, but even to a 2038 or a later deadline.

Around 10 years ago, dollar/rupee was in the early 40s, but let us say 39 as Marc says. Today it is touching 73. That is almost a 6.5 percent depreciation yearly. At this rate it could hit 100 in 2023-24, not even 2028 as Marc claims. Yet my claim is that the rupee will never hit triple digits – or to be more technically correct, such a scenario is extremely unlikely. This seems ludicrous yet the ludicrous is logical. To explain my hypothesis better, we need to quickly explore a few building blocks. First we need to understand a few concepts: nominal exchange rate, real exchange rate, purchasing power parity, the Balassa Samuelson effect and economic convergence. Finally we must ask: OK, but why now and why India?

The nominal exchange rate is what we hear all the time: for example, 70 rupees being exchanged for one US dollar in the market. Assume that interest rates in US and India are 2 and 10 percent respectively. Finally, consider the scenario that the nominal rate stays at 70 for over a year. Now an Indian who had 70,000 rupees in her account last year will this year have 77,000 rupees. An American who had 1,000 dollars in her account will have 1,020 dollars. But if the Indian travels to the US, she can get 1,100 dollars after conversion! How is the Indian suddenly better off by 100 dollars while the American is only by 20? It is because the nominal exchange rate did not depreciate by around 8 percent (10-2 percent) i.e. the rupee did not go down to approximately 75.6 from 70. Why are interest rates so different to begin with? Largely but not entirely because inflation rates are so different.

The higher-inflation currency should generally depreciate by around the inflation differential in nominal terms for the real exchange rate between the two currencies to remain constant. If the nominal exchange rate remains same, then the higher-inflation currency has actually appreciated in real terms. Also, a currency’s real value is not just measured against only one other currency but a basket of currencies, with the various pairs of real exchange rates being standardized at the starting year (say, at 100) and their weights depending on the ongoing trade between the countries. This resultant number is called the real effective exchange rate or REER of a country, and depending on the number of trading partners in the basket the REER can be narrow or broad; generally, the broader the index the more useful it is. Now as Figure 1 shows, in the last 25 years or so, Indian and American REER have largely remained constant or appreciated very marginally but the Chinese REER has almost doubled. Why is that?

Figure 1: REER for India, China and USA [Federal Reserve Bank of St. Louis, additional analysis]To better understand how a REER is supposed to move over time once we ignore the statistical noise, it is important to consider the concept of relative purchasing power parity (PPP). A basket of similar quality goods and services that may cost 15 lakh rupees in a Tier-2 town in India (~$20,000 at market exchange rates) will probably cost 65,000 to 85,000 dollars in a Tier-2 town in the United States. That is, the cost of living in the United States is 3.5-4x more expensive. Of course, finding the same basket of goods in two countries can be difficult given different preferences, substitution effects and varying public goods. Nonetheless, we have data from the World Bank, other organizations, and almost real time data from portals such as Numbeo that compare cost of living (Numbeo data is urban-biased, so if anything it should underestimate the multiplier for India; on the other hand many goods and services are not of the same quality in India as they are in the US). The multiplier for US vs China is ~2x and for Europe/Australia/Japan it is around the same cost of living as the United States. Mapping incomes and PPP multipliers shows a fairly tight relationship as can be seen in Figure 2 (the R-squared valued for the trend line is 89% which is relatively high.) Hence when India is also as rich (or poor) compared to the US as China is today, its multiplier should also be around 2, probably less. We will get back to this shortly.

Figure 2: GDP converges with GDP PPP as incomes rise [Source: World Bank, additional analysis]But why is the cost of living different across the world in the first place? It is because even as many goods have become globally tradable with only transportation, tariffs and taxes being the real difference in final prices, many services must perforce be local. It is difficult to get a haircut ‘made in China’, at least as of now (in a few decades, with revolutionary technologies such as augmented/virtual reality, robotics and faster broadband this too may change accelerating global convergence; for more on this please read Richard Baldwin). So even though your iPhone may be similarly priced in India and Norway, the haircut may be up to 50x dearer in the latter. Same for health, education, whether paid for through fees or taxes, and other professional services. Hence, as a country ‘X’ catches up with the United States, the multiplier of the US cost of living compared to X goes down, eventually to around 1. This observation, stated in different ways, is known as the above-mentioned Balassa-Samuelson effect. Some approximate numbers: India’s PPP GDP per person is $8k, but its market GDP per person is $2k. Similarly, China’s are around $18k and $9k respectively. The US’ are $62k and $62k. We can see the PPP multipliers here: around 4, 2 and 1 respectively. As a country becomes relatively richer its "multiplier" falls and REER appreciates, on average.

Now, back to Figure 2. In real PPP $ per capita terms, American incomes are increasing by ~2% and Indian by ~7%. India’s current PPP incomes is ~8x less than America’s. At a 5% convergence then (conservatively speaking), in 14 years, India’s PPP incomes will be 4x less than the US, or 25% of the US. Figure 2 shows that the “multiplier” then should go from around ~3.5x today to at least 2x. That is a 4% real appreciation of the rupee annually! Let us be even more conservative and assume that is just 3%.

Earlier, when inflation was in almost double digits, say 7-8%, then the differential with US inflation was 5-6%, and the rupee would depreciate nominally by say 5% every year. Then "just" the 0.5% average annual real rupee appreciation was because while India was getting richer, it was not doing that rapidly as population growth was also high, foreign capital was still apprehensive and domestic policies were also hesitant in pushing for growth and infrastructure aggressively, some exceptions notwithstanding. Now with a 4% inflation target in India (2% to 6% range), a target which has been credibly met over the last few years in trying circumstances, the inflation differential has gone down to ~2%. But again, let us be conservative and assume that Indian inflation will hover around 5% and not 4%, therefore difference with US inflation will be 3%. Along with the REER appreciation calculated above conservatively (3% in the next decade or so), going forward the nominal exchange rate of the rupee should remain flat, say around 65-80 rupees for the next 10-15 years at least. Hence, the rupee could remain stable around 70 for a long time to come. Of course, after a few decades you lose out the benefits of REER appreciation as incomes and costs start becoming comparable with the large “technology frontier” economy aka the US for now, however by then the inflation target is also likely to be brought down (in China, the inflation target, albeit a less formal one than in India, is 3% - halfway between India and the United States.)

This is indeed a radical claim – that the rupee will never hit triple digits. But the logic of real appreciation due to rapid productivity growth in India while nominal appreciation stopping if not reversing due to a lower inflation regime in India is inexorable. Just because we are in a strong-dollar, strong-oil phase temporarily does not mean that this situation will last.

Indeed, if real Indian per-capita growth is even faster than 7%, as I think will be as India’s dependency ratio falls, labor force participation rises for women, the 4% inflation target is met and reform momentum is at least somewhat continued - then even in nominal terms, the rupee could actually appreciate and not insignificantly. This should not be so surprising. From 2005 to 2013, USD-CNY went from 8.3 to 6.1 (and is currently at 6.9). India today is approximately at a similar stage of development and demographic profile where China was in the early 2000s, and with the recent focus on infrastructure as well as ending inverted tariffs, Indian manufacturing exports could well take off just as Chinese supply chains get partially dis-intermediated by Trumpian tariffs. Then even as the Indian workforce gets more educated than ever, our service exports might witness the additional benefits of a new and larger boom in outsourcing thanks to 5G, teleporting and related technologies even as similar developments in artificial intelligence and robotics might take longer.

Moreover, the acceleration in hybrid and electric car/bus adoption globally along with the construction of new pipelines in core American shale producing areas mean that the current crude rally may have only a year or so left, and that from the 2020s onwards “peak demand” could be much more likely than “peak supply” - though the variables are too many and too complicated interlinked. Finally, even as India’s broad REER gradually rises, the dollar itself is due for a partial correction given that the dollar and American markets are overstretched. The US is out of fiscal and monetary bullets, and while the party may last for some time, but the emerging markets trade may soon start looking like a smart contra bet. And the next EM five-year boom could make the 2003 to 2008 bull run look like peanuts.

However, even if all these optimistic scenarios do not fully play out, at the very least the rupee is likely to maintain its current nominal trading range with respect to the dollar, and is highly unlikely to ever touch 100. This will of course have massive investment and policy implications. Policy wise, this hypothesis should not be taken as a signal for complacency and indeed when the next time FII flows return we should build stronger reserves. REER appreciation is a fundamentally correct story and will sooner or later happen strongly, but what Keynes said, "the market can remain irrational longer than you can remain solvent", applies equally to many emerging countries as well in their early stages of development. Had the RBI built stronger reserves and let the nominal rupee exchange rate depreciate marginally more in the last few years, a situation would not have arisen whereby we have real yields as high as 4% (10-year sovereign minus CPI) and a fundamentally undervalued currency.

Harsh Gupta, CFA is the Chief Investment Officer of Ashika Investment Managers and is the co-author of Derivative (Cambridge University Press).