The underpinnings of RBI monetary policy
The third bi-monthly monetary policy decision on 2 August to reduce the policy repo rate by 25 basis points was widely anticipated. I am not going to join the feverish discussion on the quantum of the repo reduction (too little, too late?), how much will be transmitted, and whether whatever is transmitted will have any impact on the real economy. While evaluating decisions of the monetary policy committee (MPC), it has to be remembered that they have been given a blinkered mandate, a single instrument (the repo rate), and a single inflation outcome scale on which they will be judged. If asked what their decision will do for the real economy, they could quite legitimately respond that the question falls outside the syllabus on the basis of which they are taking their bi-monthly exams.
There is one issue, though, that falls right within their syllabus. That has to do with their projections of inflation as measured by the consumer price index (CPI), the officially named indicator. As every central bank of an inflation-targeting country will tell you, monetary policy has to go by inflation projections rather than past or even current rates of inflation.
Criticisms of the inflation projections of the MPC have been voiced vigorously in many quarters, including by the chief economic adviser. What people are looking for is not infallibility, since projections are conditioned by such information as is available at the time, and cannot get what will happen in the event perfectly right. But what is necessary is some cognizance of the criticality of getting inflation projections right, as a necessary condition for getting monetary policy right.
The previous (second bi-monthly) monetary policy statement of 7 June had at hand only the inflation rate for April, which at 2.99% showed a sharp decline, relative to March. Based on that, the statement concluded (para 14): “If the configurations evident in April are sustained, then absent policy interventions, headline inflation is projected in the range of 2.0-3.5 per cent in the first half of the year and 3.5-4.5 per cent in the second half.” The fan chart showed a mean expected inflation rate of 2.4% for Q1 (April-June), and 2.8% for Q2 (July-September).
As we know now, the inflation rate for Q1 printed at 2.2%. The latest statement of 2 August acknowledges this softening of inflation relative to expectations (para 19): “The MPC noted that some of the upside risks to inflation have either reduced or not materialized…”. However, the fan chart in the new policy shows the expected rate of inflation for Q2, the next quarter, at 2.8%, exactly where it was in the June chart. So the softening of inflation in Q1 relative to expectations did not change the projections for Q2.
The Q3 and Q4 projections are more curious still. Projected inflation in the June policy for Q3 and Q4 stood at 3.6% and 4%, respectively. In the latest policy, the Q3 and Q4 projected rates have actually risen to 4% and 4.4%, respectively. These projections are supposed to include the impact of house rent allowance (HRA) under the seventh pay commission awards. But there are several references in the statement to projections excluding HRA. If HRA makes such a material difference, how about a chart showing the difference to support the text?
If the upside risks to inflation as earlier visualized failed to materialize, and were indeed the basis for the reduction in the repo rate by 25 basis points as stated in para 19 of the latest policy, why then do the inflation projections in the fan chart show enhanced expectations for Q3 and Q4?
If the uncertainties surrounding inflation projections as listed in para 16 are such that their net impact actually pushes up expectations for the second half of the year as shown in the fan chart, why was the repo rate reduced?
When the voting MPC was set up, it was tempting to sit back and leave the driving to them. I still prefer not to question their judgement, but there has to be a clear link between inflation projections and their policy rate decisions.
On a related issue, I was somewhat disconcerted by the divergent vote of one of the three RBI (Reserve Bank of India) members of the MPC. The uniform voting pattern across MPC members was initially a matter of concern, and the emergence of dissent in the June policy by an external member was encouraging. But this new evidence of dissension within RBI is worrying. If their differences are centred around the inflation projection model, that is very worrying indeed.
Any inflation model has two basic components. There are the endogenous elements in the model, which power the internal dynamics of price movements. Then there are the exogenous elements that have to be fed in as and when they happen and become known, like acreage sown, rainfall, trade policy (for foodstuffs), international commodity prices and exchange rate movements.
It is these that the committee probably just has not left enough room for. There are multiple sources from which crop intelligence is obtainable on a weekly basis. There is the directorate of economics and statistics in the ministry of agriculture which is supposed to perform this very function. The RBI itself has a vast network of state-level offices, which could have regular liaison with state agriculture departments and directly relay market intelligence reports to the head office.
The problem is that our inflation-projection models follow those in the US or other developed countries, with excessive reliance on formal data, which in those countries are available at the national level in real time. This is India. Without citing Indian exceptionalism, we are more regionally diverse, and less supplied with a formal real-time data base. Informal channels of information flow have to be tapped to get anything like a reliable handle on inflation projections.
An even more serious problem than inflation projections is about the direction of credit. No matter what happens to repo and lending rates, the credit needs of small-scale and medium-size enterprises (SSMEs) are being systematically neglected. Even at existing lending rates, there is a vast unmet need for formal credit from that sector. The poor credit offtake, which is mentioned in every monetary policy statement, is a result in large part of neglect of small borrowers. Historically, SSME credit needs have been met through the informal channel, where rates are astronomically high.
Even after re-monetization, their pre-demonetization cash channels of credit might still not be fully restored. If there is no replacement of those informal channels with formal bank credit, there will be disaster on the growth and employment fronts. We have priority sector credit mandates in place, but the banking sector views those as a burden to be accomplished with minimal effort at due diligence. The Jan Dhan initiative will remain sadly incomplete if the induction of small savings into formal channels is not accompanied by a corresponding flow of credit to the small borrower.
Luckily, the RBI recognizes this problem, and convened a conference on the subject at the College of Agricultural Banking in Pune at the end of June. Deputy governor S.S. Mundhra (just retired) in his address to the conference (available on the website) captures succinctly the reasons for what he terms “excessive lending to the corporate sector”, where with little effort one could create large credit volumes. “Creating similar volumes in the priority sector would have required commitment of larger resources in terms of branch staff.”
There are quite simply not enough trained banking staff in place with expertise in risk assessment for small loans. There is not enough tapping of expertise available at the Institute of Rural Management, Anand (IRMA) and other such institutions on bringing the credit-worthy small-scale entrepreneur into the formal credit fold. The MUDRA initiative merely re-finances loans. The loans themselves have to originate at the point of the credit need.
The Mundhra lecture identifies a number of organizational infirmities within banks which limit access for the small borrower, including their credit scoring systems. Branch reach is apparently not a limitation, what with data for 2016-17 showing that public sector banks could not achieve the sub-target for micro enterprises (8% within the overall priority sector target), where private sector banks which have fewer branches were able to exceed their target.
He also refers to data showing that the maximum priority sector credit flow has gone to financial intermediaries like non-bank finance companies, micro-lending institutions and housing finance companies for on-lending. In itself this is not such a bad thing, but a welcome RBI initiative to allow co-origination, whereby banks join these downstream financial intermediaries in an agreed structure of risk sharing and covenants around the loan, could improve terms for borrowers.
The debate around monetary policy cannot stop with the repo rate, and rate transmission issues. It has to go beyond into whether credit is reaching the last mile, in a country as dependent as India on the informal sector to provide employment to the annual flood of entrants into the labour force.
Finally, the excellent statement on regulatory policies accompanying the policy statement shows that the marginal cost of funds-based lending rate (MCLR) introduced in April 2016 has not displaced the base rate, which continues alongside. An explanatory note from the banking regulator on how the credit portfolio of commercial banks is supposed to be split between those two benchmarks would be most helpful to current and potential borrowers.
Indira Rajaraman is an economist.
Comments are welcome at firstname.lastname@example.org