03 September 2018

Does Public Debt Crowd Out Corporate Investment?


Ugo Panizza, Professor of International Economics and Pictet Chair in Finance and Development at the Graduate Institute, and coauthors Yi Huang and Richard Varghese (Assistant Professor and PhD candidate at the Institute, respectively) have published a paper on the crowding out of corporate investment by public debt. They find that this is particularly costly for those firms that rely on external financing and thus, are likely to be credit-constrained. More details with Professor Panizza.

What gave you the idea to write this paper?

People have been worrying a lot about the increase in public debt and there is a literature that shows that higher levels of public debt may be bad for economic growth. There is in fact a strong negative correlation between the two. However, the Macroeconomics literature has not been successful in identifying a causal effect. It may be that the low growth-high debt relationship exists simply because when you use the debt-to-GDP ratio and the denominator goes down, the debt-to-GDP ratio goes up. This may be due to Keynesian policies – you increase public deficits when the economy goes into recession, so in this case it is low economic growth that causes high debt. In the past, I have done work on this with Andrea Presbitero, using macrodata and an IV (instrumental variables) strategy… but identifying a causal link has been challenging in general.

Can you describe the methods you use in the paper, and your main findings?

We use microlevel data to try to address this issue. We are able to include country-year fixed effects, which capture all the possible shocks that could affect economic growth and debt at the same time. Further, we look for a specific mechanism through which debt may affect growth, i.e. reduce the ability of firms to access credit. We basically test whether higher levels of public debt tighten the credit constraints faced by firms – specifically those firms that are more likely to be credit-constrained to start with because they are smaller, younger, without a well-established credit history and not listed in the stock market. We find that, indeed, higher levels of debt tend to tighten the credit constraints of such firms.

To elaborate a bit, a standard test of credit constraints begins with the fact that in a perfect world, firms can access all the capital they want. Their internal resources – the cash they generate – should not be related to investment. If I have a fantastic idea, I can walk to a bank and borrow money. If markets work, I will get a loan and make my investment. However, if markets don’t work, banks will not be there to lend me money for my fantastic idea. So, I will be able to invest only if I have the cash. A typical test of credit constraints is therefore based on the correlation between investment and cash flow, i.e., how much cash the firm has. The idea is that the lower the credit constraints are, the lower the correlation. So what we do is test whether the level of debt increases this correlation between cash flow and investment.

Is the relation between public debt and corporate investments working through higher interest rates or reduced availability of credit?

This is interesting, but it is something we cannot test. We do not have good data on the interest rates paid by the firms in the sample. The most basic idea from Introductory Macroeconomics is the IS–LM model where the crowding out goes through interest rates. So if you increase public expenditure, this increases the interest rate, which reduces investment simply because it becomes more expensive. But there could also be a quantity effect. In a world where people are credit-rationed, when the government borrows more and there’s only a limited amount of funds to lend, the banks will lend to the government, as it is a safer borrower. In this case the interest rate will not respond – as it is not in equilibrium due to the presence of rationing. Now, it would be nice to test this but we would need to have information on the interest rates paid by the individual firms, which we simply do not have.

What are the policy implications of your paper – and are they different for emerging and advanced economies?

The policy implication is that there might be costs linked to increasing debt. Now, we’re not taking a strong stand on this because there might also be benefits in the sense that if you are under a recession, it might be ideal to run a deficit and increase debt to stabilise the economy. But we identify one cost, which is the crowding out of private investment. In the paper, we don’t find big differences between advanced and emerging economies.

What potential do you see for future research on this topic?

This paper is linked to and follows up on another paper we wrote, using the same methodology to look at local government debt in China. We wanted to see if higher local government debt is correlated to lower private investment in Chinese cities. So, one potential area is to go into a more disaggregated level and look within countries. This of course means you have to consider big countries, apart from China. Among emerging markets this includes India and Brazil, and among advanced economies, the United States. They would all be interesting cases.



Full citation of the paper:
Huang, Yi, Ugo Panizza, and Richard Varghese. “Does Public Debt Crowd Out Corporate Investment?” International Economics Department Working Paper HEIDWP08-2018, The Graduate Institute, Geneva, 2018.

Interview by Nayantara Sarma, PhD Candidate in Development Economics.
Front picture: Currencies by 16:9clue, licensed under CC BY 2.0.