Kahneman’s System-I, Economic Survey and Macro-101

Economic Survey – the flagship publication on Indian economy released a day in advance of budget presentation is in its seventh year of two volume release. Volume-I is more about ideas, issues, debates. Volume-II is the usual economic round up. Volume-I gives an intellectual foundation to various reform agenda/action points those are either being debated in public or raise new issues encouraging more constructive dialogue.

Volume I is also for university students and teachers given the technicalities to which it has gone into. It is not a simple narrative. It has hypotheses, theory, empirics, and a little bit of latin characters as well.

Current Survey has a chapter titled “Does Growth lead to Debt Sustainability? Yes, But Not Vice- Versa!”. It says:

 …. the Survey endeavors to provide the intellectual anchor for the government to be more relaxed about debt and fiscal spending during a growth slowdown or an economic crisis.

When COVID pandemic hit the World and India as well, there was an increased need to augment public spending. Fiscal deficit of India in 2019-20 was 4.6% i.e., INR 9.3 trillion. Govt. budgeted in 2020-21 to bring it down to 3.5% of GDP to INR 7.9 trillion. Well, what we ultimately ended up having is a 9.5% of fiscal deficit for the year 2020-21 amounting to INR 18.48 trillion. Not only that – the primary deficit was 5.9% of GDP – a number that really bothers economists. We are still not out of the wood. Govt.’s budgeted fiscal deficit for the year 2021-22 is INR 15.06 trillion i.e., 6.8% of GDP. Many argue that it is still not enough to bring the economy on the track – more expenditure is needed from government.

At the same time, there are counter arguments that – if we increase our fiscal deficit, there is a risk that our sovereign ratings may be impacted, we may get into a fiscal mess, debt sustainability may be questionable etc. That is the background in which the Survey presented this chapter offering an intellectual backing to increased govt. expenditure. The chapter is 41 pages long with ~13786 words. If we have to summarize these 13786 words into 280-character limit of this Larry T bird, it will go like this:

 India’s growth prospects remain bright. We have much unutilized resources where we can put our borrowed money. As long as our GDP growth is higher than the rate at which we pay our interest, our debt would remain sustainable. We will never get entangled in a labyrinth. [270-character count with spaces]

Read the original chapter to see how the arguments are built on the basis of theory, empirics, graphs, charts etc. I am not going to discuss the entire chapter. I am getting back to my original System-I thinking.

System-I Thinking and Economic Survey

 The survey stated that –

 …It is a call to break the intellectual anchoring that has created an asymmetric bias against fiscal policy. Once growth picks up in a sustainable manner, it will be the time for fiscal consolidation. But, for now, fiscal policy will have to remain centre-stage to support growth in the foreseeable future.

What is this asymmetric bias against the fiscal policy? There are many. One such intellectual bias which the Survey may be alluding to is ‘CROWDING OUT’. You would have heard something like this – if government borrowing increases, it would crowd out private investment.

How?

Well, there is a fixed size of loanable funds market. Even if it is not fully fixed, it is not infinitely elastic. So, if the government borrowing increases, it would add pressure on interest rate. If the interest rate increases, private investors would find it unattractive to borrow and invest. That is how government borrowing would create more mess in the economy than really helping the economy.

In page number 64-69, Survey has exclusively focused on this question under the section - Crowding out due to public expenditure?

Well, you would expect the Survey to answer that question with a big NO. You are right. That is what the survey has answered. But the Survey has not answered that on the basis of gut feelings or intuition. It has quoted many empirical literature, theoretical arguments, and presented graphs/charts supporting its claim.

Let’s see the figures that Survey has put in that chapter in supporting the argument that there is no crowding out of private investment due to increased public spending.

Figure-A plots the change in public investment vis-à-vis change in private investment. It is for the period FY-1951 to FY-1990 – our pre-liberalisation era. The correlation is negative and statistically significant. The survey explained this graph saying – “…during the pre-liberalisation period of FY 1951-FY 1990, a negative correlation between changes in public investment and changes in private investment provides evidence consistent with the rationale of fixed loanable funds and possible crowding out.”

The survey is incredibly careful in using the word "may" liberally throughout the chapter. It has used the verb 'may' at least 27 times against 11 times of 'will'. As we know 'may' is in the realm of 'possibility'. When we say it 'may' be true, what it also implicitly implies that under certain conditions, it may be untrue as well. That is what Macroeconomics all about. You make inferences on the basis of limited information being fully aware that there may be another story/another side.

A little bit of further digression here. I normally avoid using adjectives while writing on a technical topic but there is a reason why I am using the word ‘incredibly’ here. Survey wrote a chapter on GDP in 2019-20. This chapter-10 opened with a quote like this –“Correlation is the basis of superstition and causation the foundation of science.”

The Survey then argued in the chapter:

The results clearly establish the concern that the correlations studied as a diagnostic for GDP growth are notoriously non-stationary: not only do they flip signs frequently over various 3-year or 5-year time periods from 1980 to 2015, their values change significantly over this time period as well.

The survey, therefore, espoused using quasi-experimental method/counterfactual methods rather than leaning on correlation as we know correlation is not causation.

Thus, Current Survey in the context of Debt Sustainability is very careful in using the verb may as framing matters a lot in macroeconomics. Survey could have undertaken a rigorous examination of this crowding out debate but that would have meant writing another chapter. In the absence of that, it has given references to other studies which are based on strong econometric foundation. But, it is also important to note that- one can always find other empirical studies with results in the opposite direction as well.

Digression over.

Let’s look at now figure B.

Figure-B is the same graph with a new time period – FY1991 to FY2019. The correlation coefficient as given in the survey as zero. I would presume it is not actually zero, it is simply statistically insignificant hence survey has put it as zero which means public investment and private investment are not correlated at all statistically.

In figure C and D, survey explores if higher government debt leads to lower corporate debt over FY 2001 to 2019? It has taken change in corporate debt in figure C and change in bank credit in figure D against change in government debt. It has found the correlations to be statistically insignificant showing no statistical association.

Taking together figure B, C and D, survey has shown that there is no relationship or no association between increase in public investment and public debt with private investment, corporate debt, and bank credit. All measured in change terms.

Therefore, if your system-I triggers when you get the H1 or H2 borrowing calendar in terms of its impact on private investment or corporate debt, take a deep breathe. Pause for a minute. Ask yourself – is the simple narrative that you have formed in your mind that government borrowing would impact private investment holds true empirically on a medium to long-term basis?

In your mind also, you have to think whether you are thinking about bond traders, stockbrokers, overnight money market or you are thinking about the Economy - the big elephant. Thinking about economy is not like a T-20 match. For economists, what happens overnight or next month is less of a concern. They think whatever action government is taking today, how it would give a shape to the economy over a medium to long-term or over a business cycle. It is up to the central bank how creatively it is managing those auctions or timing it so that disruption is minimal. T C A wrote an article in Business Standard a few days ago with the title – The discourse on the economy has been hijacked by the bond market types. Well, that is the style of T C A Srinivasa-Raghavan but there is a grain of truth also. You would have seen the curiosity and press coverage in terms of what is happening to high frequency indicators – is the economy back to track? What happened to x indicator, y indicator, z indicator? Well, it makes sense for traders to track those but not certainly for policy makers or economists who see the Economy as a big Titanic in an open sea. There will be short-run adjustment and ripple effects of certain actions but ultimately the ship would take turn to a desired direction if the prognosis was correct as well as timing of the interventions.

Econ-101 of Crowding out effect

Let’s get back to the last bit of the Title that is Econ-101. What really economic theory tells us about crowding out effect?

Let’s start with two basic macroeconomic identities.

You would have seen in GDP press release that GDP is presented under the heads of Consumption (C), Investment (I), Government Expenditure (G), Exports (X), Imports (M). Let’s leave aside valuables and inventory etc. We can write this as:

Y=C+I+G+[X-M] …. Identity 1

 Here Y is your GDP.

When you get your salary, what would you do? You consume or pay tax or save. Let’s write that:

 Y= C + S + T…. Identity 2

 Now if you equate these two identities, we will get something like this:

 G-T=[S-I] +[M-X] …. Equation 3

Here G-T is the budget deficit. Equation 3 tells us that government budget deficit is either could be financed by running perpetual current account deficit or by increasing savings or cutting down investment.

Let’s now imagine that running a perpetual and rising current account deficit is not an option as it has various other implications. Let’s for the time being, delete M-X from the equation. Then what is left is:

 G-T=S-I…. Equation 4

To feed the rising deficit, either we need to increase our domestic savings or private sector has to cut down its investment. It operates like this. Govt. to support its increased borrowing program would auction bond through RBI (unless it monetizes which has its own implications). When the supply of bond increases in the market, price of the bond would fall. We know the bond price and interest rate operate in reverse direction. When the price of bond falls, interest rate would increase. When the interest rate increases, it would impact the entire yield curve. Many private investors would find it infeasible to borrow at such higher interest rate as their business model would not support this rising financing cost. Private investment would go down.

This is the textbook explanation of crowding out effect.

There is another phenomenon known as Ricardian equivalence. Ricardian equivalence shows how Keynesian prescription of stimulating economy through public expenditure fails. For example – assume that govt. has increased the spending through borrowing. People know that money does not grow in tree and at some point, govt. would increase the tax to collect all the money that is required to payback the outstanding debt. Therefore, people would not increase their expenditure in response to increased public spending. They will rather save for bad days. Keynesian multiplier would fail to operate and rising public spending will have no impact on the economy. Economic Survey gives numerous explanation - why Ricardian equivalence does not hold both conceptually as well as empirically.

I also think that the Ricardian Equivalence is a little bit farfetched. Do people really care about the next generation or do they really hold back their consumption thinking at an uncertain future time, government would increase the tax rate so save money now? I do not think any normal individual think in that line. While Ricardian equivalence is quite elegant in its logic – it has serious issues in its assumptions as well as practical applications.

But, let’s get back to the first part of this Econ-101.

G-T=S-I…. Equation 4

Why then in the presence of govt. borrowing, private investment would not fall? There could be multiple explanation for this – one is complementarity. Public investment in many cases help the private sector to invest. It creates roads, ports, bridges, cold storage, power generation and distribution facility etc. Those actually work as complementary factors and help private sector investment. Second, when the economy is operating below potential or there are unutilized resources, public expenditure would not only create assets, it would also generate jobs and incomes. As job and income increases, private sector savings would also increase. Corporates would also get encouraged to mobilise their idle cash balances and be more proactive in their investment decisions. A virtuous cycle would kick-start like this: Public Investment-Jobs/Income-More demand- More private sector investment – more jobs/income- more savings- more investment. It would go on.

Question therefore arises – if the public spending is so innocuous in its implications, then why there is so much noise around? Well, nothing comes as a free lunch. Loanable funds market would not increase overnight therefore one can see some strain in the bond market, overnight money market. Banks whose Asset-liability or infrastructure companies whose books are based on the yield curve would also feel the strain. Those short-term pains need to be creatively managed. Moreover, when you are thinking about the economy, you cannot always think it like T-20 match. T-20 has its charm but who does not enjoy a good 50-50 or test match!!!

 ______________

[Note: While the survey has not uttered once in this chapter about Modern Monetary Theory, there are many strands that closely match with the MMT argument. Do you notice that? More so on that as well as on debt sustainability.]