The monetary dimensions of inflation
Addressing inflation requires a deep understanding of how currency moves from banks into the cash economy
The empirical facts of the inflationary episode are well known and extensively documented (see for example, Dev, Pandey and Rehman 2015), and there is little need to cover the ground yet again. Suffice it to say that inflation measured by the Consumer Price Index (CPI) remained at or near double-digit levels for five consecutive years (2009-10 to 2014-15) despite a consistently tight monetary policy followed by the Reserve Bank of India (RBI) throughout the entire period.
In mid-2014, the RBI announced its adoption of inflation targeting (IT), and proposed a “glide path” for CPI inflation to go down to 8% in January 2015 and then further to 6% in January 2016. To everyone’s surprise, the inflation rate plummeted shortly thereafter and went below the January 2016 target by December 2014.
All commentators are agreed on one thing: that the inflation was led by food prices. There are of course quite expectedly differences on the reasons underlying the food price inflation and its persistence. However, nobody appears to have a credible explanation for its dramatic decline. Moreover, all the explanations of food inflation appeal to real factors – the so-called structural causes – and there is no mention whatsoever of any monetary mechanism behind the propagation and perpetuation of the inflationary process.
This is surprising in view of Nobel Laureate Milton Friedman’s dictum: “Inflation is always and everywhere a monetary phenomenon, in the sense that it cannot occur without a more rapid increase in the quantity of money than in output.” Strangely, even the RBI, which is the designated anti-inflation fighter and custodian of monetary policy in India, has not articulated any monetary explanation of what has happened. Its position has been that the inflation has been caused by supply-side factors over which it has little control, and its contractionary monetary stance has been justified on the grounds that it is necessary to control inflationary expectations and prevent further acceleration of the inflationary process.
In this article, I argue that while the initiation of the inflationary process may have been rooted in structural factors, there is a monetary story to be told about its propagation across commodities and perpetuation over time. This understanding explains RBI’s helplessness and may help in designing appropriate policy interventions for similar inflationary episodes in the future.
Monetary dualism
The notion of ‘dualism’—that a country can be viewed as having two distinct sectors—has a long and rich history both in development economics as well as in popular Indian discourse (this is referred to as the “Bharat-India” divide). I propose a specific form of dualism: that the Indian economy is composed of two sub-economies—a credit economy and a cash economy (roughly defined as the entire rural economy plus a significant part of the urban informal sector). In the credit economy, all transactions within it are mediated through the banking system; whereas in the cash economy, all transactions are in cash. The key assumption is that all transactions between these two sub-economies are carried out only in cash. I further assume, following Friedman, that the average price level of goods produced by the cash economy is determined by the amount of liquidity or the stock of currency available in the cash economy—along with other factors affecting demand.
The total stock of currency in the country is used for three purposes: (a) transactions in the cash economy; (b) transactions in “black” assets; and (c) reserves held by the banking sector (including bankers’ deposits with RBI). The total stock of currency is determined uniquely by the RBI as a policy decision. There are two predominant instruments through which the RBI can increase the stock of currency in the country—purchase of either government securities or of foreign exchange. In both cases, RBI action directly increases only the currency reserves of banks, and not the currency held by the public. It is, therefore, important to understand the mechanisms by which currency moves from the reserves of the banks to the cash economy.
There are three main channels. The first is the net value of transactions between the cash and the credit economies. If the money value of purchases by the credit economy from the cash economy exceeds the value of purchases by the latter from the former, the difference between the two values will be reflected in an increase in the stock of currency in the cash economy. The second channel is through the government’s fiscal activities. The government not only buys and sells goods and services with the cash economy (which are included in the first channel), but also makes substantial transfer payments in the form of social security, scholarships and the like. The third channel is the net value of cash injections into the black economy less the cash withdrawals through money laundering.
Effects of an agricultural shock
In such an economy, if the production in the cash economy is adversely affected by a shock (say, a monsoon failure), the prices of agricultural products will rise since the liquidity available will now exceed the availability of goods. If we further assume that food (a major component of agricultural output) is an essential good so that its demand in the credit economy is inelastic (not responsive to changes in prices), then consumers in the credit economy will withdraw currency from their banks for purchase of food. This will increase the supply of liquidity in the cash economy, leading to a further increase in prices.
Even when production returns to normal (say during the next harvest), food prices will not decline to the original levels since there would now be more liquidity in the cash economy than earlier. This permanent increase in the price of food relative to all other goods will continue to induce further flows of cash from the credit economy into the cash economy, leading to further increases in food prices. This process will continue until such time as the induced shift in demand towards credit economy goods eventually leads to a zero net transfer of currency between the two sub-economies. Such a shift in demand will occur as the result of the price of credit economy goods falling relative to that of the cash economy goods.
In the credit economy, so long as the currency reserves held by banks exceeds the minimum levels required for precautionary purposes, there will be no reduction of credit available. Thus, there can be a fairly extended period of food inflation with no change whatsoever in any of the monetary aggregates (M0 or M3). The inflationary process in such a case is supported by a compositional change in the holdings of currency from passive bank reserves to active circulation in the cash economy.
The role of minimum support prices
It has been forcefully argued with empirical evidence by Surjit Bhalla and others that the principal determinant of food inflation in India has been increases in the minimum support price (MSP) of cereals. The argument is that the MSP not only sets the floor price of cereals in the country but also affects the prices of other agricultural products by inducing a shift of land towards cereal production, thereby reducing the availability of other agricultural products. While the floor price argument is undoubtedly true, the land-switching effect is not borne out by data on area under crops. However, the story is more complex than that, and involves the entire chain of interventions by the government in the food economy.
Every year, during harvest time, the government procures nearly 25% of the cereal production at the MSP; the remaining 75% of the grain is sold to traders, who are a part of the cash economy for the most part (in recent years, the annual procurement has averaged 60 million tonnes out of a total production of about 240 million tonnes).Thus, twice a year, within a very short time, there is a massive injection of currency into the cash economy through the procurement process.
During the course of the year, the government releases a major part of the grains procured at highly subsidised rates through the public distribution system (PDS) and a smaller part free of cost for certain government schemes. Nevertheless, this usually leaves a significant amount of grains undistributed. (Of the 60 million tonnes procured, 45 million tonnes are released through the PDS and 5 million tonnes are provided free to schemes, leaving about 10 million tonnes undistributed. The undistributed amount can either be issued as ad hoc allocation to certain states and for exports or can remain in the government’s buffer stocks.)
Thus, on the supply of cereals, the procurement-cum-PDS operation is equivalent to a mild supply shock with one important provision: there is no loss of income for the farmers. On the monetary side, these operations inject a huge amount of net liquidity into the cash economy. Over the past several years, the procurement price has been increased every year, but the PDS issue price has remained constant. Thus, the monetary injection has steadily increased year by year. Back-of-the-envelope calculations suggest that the net liquidity injection through this process accounts for roughly half of the total increase in the currency held by the public in any given year.
The inflationary consequences of these two effects can be very substantial indeed even in a normal year. Since there is no loss of income for the farmers, the additional liquidity is spent on other goods, including non-cereal food items, thereby raising their prices. Of course, a part is also spent on credit economy goods thereby reducing some of the extra liquidity, but given that the consumption basket of farmers is dominated by food, this effect is probably small (typically, food accounts for about 60% of the rural consumption basket, and about the same for the urban poor as well).
The consequence is that the price impulse spreads rapidly from food grains to all other food items as well. In so far as the credit economy is concerned, the general increase in food prices will induce a flow of cash out of bank reserves into the cash economy, thereby perpetuating the inflationary process. The larger the increase in MSP, the stronger will be this inflationary process not only because of the direct effect on food grain prices, but also because it would induce a higher proportion of food grains being offered for procurement.
What happened in 2014?
I believe that the monetary process described above is the root cause of the persistent food inflation observed since 2005 when sizable increases in MSP were made by the UPA (United Progressive Alliance) government. This was exacerbated from 2008 when the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) was rolled out nationally, which represented a new source of liquidity injection into the cash economy.
On average, MGNREGS injected about Rs.320 billion annually. As a result, in 2008-09, although reserve money grew at only 6.4%, currency with the public grew at 17.1%. Matters came to a head in 2009 when India experienced the worst drought in 30 years. This not only involved a substantial supply shock, but MGNREGS payments shot up by nearly 50% (MGNREGS pay-out in 2009 was nearly Rs.480 billion). To make things worse, there was gross mismanagement of the food economy by the government with food grain stocks actually increasing due to releases being significantly lower than the procurement during the year. This not only meant a supply shortage larger than just the loss of production, but also additional liquidity injection which further supported the inflationary process.
The inflationary impetus provided by the events of 2009 persisted for the next four plus years and eventually spilled over into the non-agricultural sector, despite the best efforts of the RBI. The RBI not only increased the policy interest rates substantially, but also reduced the rate of growth of reserve money significantly especially in 2011-12 and 2012-13 when it was 3.7% and 6.1% respectively. Currency with the public, however, grew at 11.2% in both years.
A possible contributory factor to the persistence, and indeed acceleration, of inflation was the collapse of transactions in the real estate sector. Since this sector represents the single largest repository of black assets, a part of the currency sequestered in property transactions was probably released into the cash economy, thereby partially off-setting the RBI’s monetary contraction.
In light of this monetary explanation of inflation, the dramatic reduction of the inflation rate in the second half of 2014 can be explained by three developments. First, the increase in MSP in that year was only 3.8%, which was well below the prevailing food grain inflation of 11%. Thus, the real liquidity injection was actually negative.
Secondly, despite the fact that 2014 witnessed a below-normal monsoon, the outlay on MGNREGS dropped sharply by nearly Rs.120 billion. Finally, and most importantly, the government released more than 6.5 million tonnes of food grains from its stock through open-market operations, which both augmented the supply of cereals as well as sucked out about Rs.80 billion in liquidity from the cash economy. This was helped by the fact that global food prices had fallen and therefore there was no incentive for traders to export the additional food grains.
These three factors together imply that the injection of liquidity into the cash economy in 2014 was more than 30% less than in previous years (the average increase in the currency with the public in the previous three years was aboutRs.900 billion). It was no wonder that the inflationary process collapsed. There are, however, two features of this reduction in inflation that need to be noted. First, it led to massive rural distress.
A sub-par monsoon is always bad news for rural India, and it was made considerably worse by the lack of livelihood support from MGNREGS. Moreover, the release of food grains from the government stocks was singularly badly timed in that it came roughly at the time of the harvest. This meant that the market prices received by the farmers were well below the level that would have been justified by the production shortfall, thus adding to their woes. Second, the RBI had absolutely no role to play in all of this.
Lessons
There is by now compelling evidence which suggests that the normal monetary transmission mechanisms either do not work or are very weak in most developing countries (Mishra, Montiel and Spilimbergo 2010). It is amply clear from the recent inflationary experience that this is true of India as well. In countries characterised by monetary dualism, such as India, this is only to be expected.
Addressing inflation which has its origins in the cash economy requires a deep understanding of the processes by which currency moves from the coffers of the banking sector into the cash economy, and what the central bank can do about it. While there is considerable research globally on the transmission mechanisms through which monetary policy, especially interest rate changes, work through the credit system, there is virtually none which address this particular issue.
Now that India has adopted inflation targeting as the centrepiece of its monetary policy, this lack of understanding can be very costly. Inflation in India can arise from a variety of causes, and it is important to know what the appropriate monetary instruments are for each case. Use of inappropriate monetary instruments can do more harm than good, and limiting the monetary authority’s instruments to a single primary one can be very dangerous indeed.
RBI can certainly be proud of finally having achieved a ‘state-of-the-art’ monetary framework. Now if only we had a state-of-the-art economy.
