Are financial savings under-reported?

If the real repo rate, the term premium and credit spread were to normalise, borrowing rates could fall by more than 2 percentage points
Neelkanth Mishra | Last Updated at April 30, 2019 21:37 IST

(This was published in the Business Standard: link)

Are we saving enough? Like it is for households, the question is an important one for the economy too. A paucity of savings would mean inadequate funds for investment, which in turn has two adverse consequences: It makes capital more expensive and also increases the economy’s dependence on foreign capital. The latter in particular makes the economy vulnerable to volatility in global capital flows.

Of greater analytical interest are financial savings by households, as they move relatively freely through the economy and are an important driver of the cost of funds; savings by private corporations and by the public sector on the other hand generally get invested within the entity.

Historically, gross household financial savings have been about a third of total savings in the economy, and for the last three decades have been between 9 per cent and 12 per cent of GDP. The drop in this ratio to 9.4 per cent in FY17 (the year of demonetisation), a two-decade low, had raised concerns on excessive consumption and the economy’s inability to finance investments. At a time when real interest rates have been so high, which generally stimulates financial saving, this was a puzzling number. The strong rebound to 11.1 per cent reported for FY18 has been relatively less discussed, as also the continuing improvement to a likely 11.3 per cent (our estimate) in FY19, a nine-year high. This is when one uses the Reserve Bank of India’s (RBI) estimation process, which, as we will see now, may be under-estimating them.

The current account is a metric for aggregate net savings in the economy: The excess of consumption over production. But calculating gross household financial savings (called just “savings” in the rest of this article) is not straightforward, even though, given that they are by definition formal, their estimation requires fewer assumptions than most other macroeconomic variables for the Indian economy. One needs to be careful about double-counting, and at the same time, be as comprehensive as possible to include all savings products.

Drilling into components of financial savings, we discover two large areas of potential under-estimation. First, the jump in mutual fund (MF) inflows: Whereas cumulative flows into mutual funds were Rs 3.5 trillion in FY17, of which about Rs 1.7 trillion would be from households, the amount mentioned in the reported statistics was just Rs 128 billion (7.5 per cent of Rs 1.7 trillion). The reported number for “shares & debentures”, of which MFs are a part, did jump to Rs 1.5 trillion in FY18, but by then, in addition to strong MF inflows, rising retail participation in stock markets had pushed assets under management of Alternative Investment Funds (AIFs) and Portfolio Management Services (PMS) higher by Rs 1.8 trillion that year. There is nearly a 1 per cent of GDP gap therefore between the reported financial savings in “shares & debentures” and what one can estimate using parallel sources like the Association of Mutual Funds of India (AMFI) and the Securities and Exchange Board of India (Sebi).

Second, inflows into small savings schemes (like in post offices, or specialised schemes for senior citizens and girl children) have jumped from about half a trillion rupees a year till FY17 to over Rs 2 trillion a year now. But the corresponding sub-segment of reported savings (claims on government) still shows an 820 billion number for FY18, nearly half of the inflows that year.

We estimate that in total the under-estimation of savings could be potentially 2 per cent of GDP in FY17 and well over 1 per cent of GDP in FY18. It is quite possible that all the above inflows have been considered for the reported statistics, and that there are some offsetting flows that we have not incorporated (the granularity of data available to a regulator or the government is much better than to us), but placing the two sets of charts next to one another makes that appear unlikely.

There is a second big concern among observers — that state and central governments are distorting the country’s finances by appropriating a larger part of savings due to delayed fiscal consolidation and large extra-budgetary spending. However, contrary to popular fears of a deterioration, public sector savings have been largely unchanged as a share of GDP for several years, as savings of non-government public entities have offset government revenue deficits. Even at a net level, that is, after adjusting for borrowings by public sector enterprises, the drain on private savings has reduced (at least till FY17, the last data available). In the last two years, quite possibly, the situation may have worsened on the extra-budgetary spending front, but this slippage is unlikely to be more than half a per cent of GDP: Far less than the improvement in household financial savings over the same period.

These are not just statistical nuances, but can have significant impact on the cost of funds in the economy. The interest rate at which an entity borrows can be split into three parts: the repo rate, the term premium, and the credit spread. The first is the rate set by the Monetary Policy Committee (MPC), the second is the difference between the yield on the 10-year government bond (the GSec yield) and the repo rate and the third is the difference between the rate at which a company borrows and the GSec yield. All three are elevated currently; the real repo rate (that is, the repo rate minus inflation), is inordinately high given the slowing economy, the term premium is close to decadal highs, and capacity issues in the financial system are keeping credit spreads high as well. If the real repo rate, the term premium and the credit spreads were to all normalise, borrowing rates could fall by more than two percentage points.

While the repo rate may drift down with inflation being persistently lower, the credit spread may not compress anytime soon, given capacity issues in the financial system. But the term premium is unjustifiably high and should fall with better clarity on the two statistics discussed above, translating to the share of financial savings appropriated by governments being significantly lower than feared and also compared to what it has been historically. Economic growth has suffered, as the government has tried to tighten fiscally, but the benefits of that have not appeared in monetary conditions that have remained too tight. As a broader theme, this further demonstrates the need for better quality data for better functioning of markets.