No one envies the MPC members for their extended three day bi-monthly meeting from June 4 to June 6. The hawks are out in droves; the doves would rather forget this spring of discontent. It was, what, just seven weeks ago, after the April 7 MPC meeting, that the cry went out from the ramparts of the RBI — no rate hikes as far as the eye can see. We all hung up our inflation boots and celebrated the oncoming summer lull. For all of two weeks.
On April 21, the same MPC that had lowered inflation forecasts for the first half of the fiscal year by 45 basis points — from a 5.1-5.6 per cent average to 4.7-5.1 per cent for fiscal year (FY) headline inflation (H1) — now revealed that in its deliberations it had actually “felt” the opposite. In particular, one MPC member, Viral Acharya, more or less declared the rebirth of accelerating inflation in India, something his colleague Michael Patra, has been warning about since he voted for a rate cut at the inaugural MPC meeting of October 2016.
Barely had the Viral shock worn off when the markets were hit by something truly viral — inflation data for April, released on May 14, indicated a 100 bp increase (acceleration) in core inflation, excluding housing, over the March y-o-y inflation of 4.4 per cent. Indeed, Acharya had hinted at this measure of inflation as a “target” for monetary policy in the minutes of the April Monetary Policy released on April 21.
There are several battles being fought here. I also think that headline inflation as a target is problematic. Indeed, my preference for an inflation target is the deflator of personal consumption expenditures (PCE), something quite common in many parts of the advanced world, especially in the most advanced economy, statistically and otherwise, the US. If the PCE were to be used (note that we don’t have to worry about excluding food, fuel, petrol, housing, vegetables or any of the other exclusions policymakers dream of), it shows the following trend for the last five fiscal years, starting 2014: 7.5 per cent, 5.3 per cent, 3.9 per cent, 3.8 per cent and 3.3 per cent (see table). Yes, 3.3 per cent in FY18, is lower by 50 bp than the demonetisation year of FY17.
We apologise, we digress. Back to impending viral inflation in India. The 100 bp increase in core inflation is only according to the RBI’s wrong definition of core — that is, excluding food and fuel. But in India, petrol (what is considered fuel in most parts of the world, and not the price of kerosene or electricity) is part of transport and communications, and as far back as December 2014, former RBI Governor Raghuram Rajan had pointed out the “havoc” that just 2.4 per cent of the consumption basket — the consumption weight of petrol — could cause. In the December 2014 Policy Statement, he stated, “Further softening of international crude prices in October eased price pressures in transport and communication”; in the December 2015 Policy Statement: “CPI inflation excluding food, fuel, petrol and diesel (emphasis added)”.
The domestic price of petrol rose by 9 per cent, y-o-y, in April 2018. Incidentally, the price of Brent crude rose by 44 per cent y-o-y. Anyway, while crude is 2.4 per cent of the overall CPI basket, it is a whopping 6.4 per cent of the basket of goods that Acharya claims should be the inflation indicator.
As he stated in the minutes of the April meeting, the one that set the cat amongst the doves, “What concerns me is that the more persistent component of headline inflation, which is ex food and fuel, and which one can consider as the ‘signal’ given its persistence, has strengthened steadily from a trough of 3.8 per cent last June to 4.4 per cent in February (excluding the estimated impact of Centre’s house rent allowance increase, that is, ex HRA). This rise has been broad-based.”
Nowhere does Acharya, or any other member of the MPC, point out that petrol should not be included in core excluding housing (CoreXh). If they were to do that, they would find that while acceleration in CoreXh inflation was 100 bp, it was some 40 bp lower for CoreXh excluding petrol (CoreXhp). How did just 2.4 per cent weight inflating at 9 per cent cause such a large difference? Very simple: A 2.4 per cent weight in overall CPI is a highly magnified 6.4 per cent in a reduced CoreXh weight of just 37 per cent of aggregate CPI. Yes, that is correct. Core CPI is 47 per cent of the basket, and housing is 10 per cent; so CoreXh is 37 per cent. Given this small size of the inflation universe, even a wiggle can cause large distortions, as we document next.
On February 1, as part of the Union budget presentation, the government strayed from its commitment to an open economy, and imposed (debatable, controversial, and unexpected) customs duties on a wide array of consumer goods. Most of us thought (including some members of the MPC) that this will have a small impact on inflation, but given the volatility in other segments, that is food, the effect will be swamped. As documented in detail by A K Bhattacharya in the Business Standard (‘Budget decision on customs duty hike to impact imports of $85 bn in a year,’ February 12), the duties were slapped on $ 85 billion of goods. For our purposes of impact on inflation, we can ignore the import of diamonds and jewelry (it may have mattered to Nirav Modi but he is absconding!) amounting to about $55 billion.
The import duty increase for most of the $30 billion of import items was 10 per cent. Total Consumption in 2017-18 was Rs 96.3 trillion (we know that from the GDP data released Thursday) and the exchange rate averaged Rs 65/$ in 2017-18, that is a total consumption of $1480 billion, making the import items directly hit by customs duty equal to about 2 per cent of total consumption.
Assume an average inflation rate of 4.5 per cent for all goods and services; if 2 per cent were to have an inflation rate of 10 per cent (as likely happened in April when customs duties were passed on to consumers — legally the import duties became effective February 2, but there are leads and lags in trade) and 98 per cent, an inflation rate of 4.5 per cent, then simple math shows that the customs duties will cause inflation to increase by 11 bp.
That is too little to worry about you say? I agree. But now look at its effect on CoreXhp (core minus housing minus petrol), an aggregation that is only 35 per cent of the total basket. In this case, 11 bp becomes 11*100/35 or 31 basis points. Which means that the acceleration in core inflation of 100 bps is now really 100-40-31 or 29 bp. Does that justify concern about hiking rates to stave off impending inflation, as some economists are arguing?
Should the RBI be unduly concerned about an 11 bps increase in headline inflation, which is its mandate? The MPC Act categorically states that it is the headline inflation, and only headline inflation, which should be the inflation target. Maybe time has come to revise the target — but please make it the personal consumption deflator? It will make the computation simpler, and prevent economists, analysts, and policymakers from poring over multiple numbers, and multiple inclusions and exclusions due to special effects (for example, the exclusion of the effects of HRA as officially done by the MPC).
One final point for keen observers of inflation, including the MPC. There are two reasons why the headline inflation target of the MPC is lower in the second half of FY2019 than the first half. First is the diminishing of the HRA effects in the second half. Second, are base effects. A reminder — the six y-o-y CPI inflation numbers from April 2017 to September 2017 were 3.0 per cent, 2.2 per cent, 1.5 per cent, 2.4 per cent, 3.3 per cent and 3.3 per cent. In other words, the worst of the base effects are yet to be registered. Economists have published their estimate of the base effect for April 2018 — widely assumed to be at least 25 bp. Which implies that there was virtually, no objectively, any genuine, demand-effect, inflation acceleration in April 2018.
- Jobs and the pre-election year
Conventional wisdom of no job growth in 2017-18 is found to be a gross under-estimate by a correct application and interpretation of household survey data…
- Scalding oil, sliding rupee
Sometimes a crisis is brought on by faulty analysis; at others, by poor politics. The proper response to a crisis, however caused, is to change…
- Pro-women, pro-poor
India may have just witnessed the best four years of inclusive growth, thanks to the Centre’s sanitation programme..