Corporate investment and monetary policy transmission in Canada

Investment matters for monetary policy

Business investment in capital is particularly important because it increases the economy’s productive capacity, enabling the economy to sustain a higher demand for goods and services. For example, Wilkins (2019) notes that investments in Canada have accounted for about 0.6 percentage points of growth in production capacity each year over the past 15 years. Given how important investments are to an economy, the influence of monetary policy on investment is of utmost importance to policy-makers.

The Bank of Canada conducts monetary policy by influencing the availability and cost of credit in the economy and, thus, Canadian firms’ incentives to undertake investment projects. For example, firms may invest more in machinery and equipment when money is less expensive to borrow.

In this note, we find that when interest rates change unexpectedly, small businesses change how much they spend on investment more than large or multinational corporations do. This is because when interest rates rise unexpectedly, small businesses may be less able to raise funds needed for investment. Large businesses, in contrast, are less likely to face such difficulties with financing.

Small firms have a bigger investment response than large firms to unexpected changes in monetary policy

We study how monetary policy affects Canadian firms of different sizes. To do this, we use the unexpected interest rate changes that happen on the Bank’s policy rate announcement dates. We call these unexpected changes monetary policy shocks. We use the unexpected interest rate changes because changes in economic conditions can influence both monetary policy and investment. Firms can form expectations about changes in monetary policy. This makes it difficult to determine whether firms are adjusting their rate of investment because of monetary policy or because of economic conditions. Surprise changes in interest rates, on the other hand, reflect monetary policy changes that were not expected given the economic conditions at the time. These surprise changes allow us to look at the impact of policy—and gauge its effectiveness—separately from the impact of other economic changes.

Chart 1a: Small firms materially change their investment rate following monetary policy shocks

Chart 1b: Investment response by large firms is muted

Note: Shaded area shows a 68% confidence band. For each quarter, a cross-section of firms is divided into small and large firms based on the size of their total assets. Firms whose total assets are greater than the 80th percentile are labelled large, and those whose total assets are less than the 60th percentile are labelled small.

Chart 1 shows how firms of different sizes respond to an unexpected 0.25 percent increase in interest rates. We observe that firms take time to digest new information and revise their investment plans. It takes about four to eight quarters after a monetary policy surprise for small firms to adjust their investment. In addition, we see that the size of adjustment is substantial for small firms, which decrease their investment rate by up to 1.5 percentage points when the average investment rate in our sample is 7.9 percent.

In contrast, the changes in investment rates made by large firms are neither statistically nor economically meaningful. One reason for this weak response might be that large firms have ample access to cash; as such, changes in financial conditions following monetary policy surprises may not materially influence their investment capabilities. It might also be that large firms do not respond to small surprises but do substantially change their investment behaviour when faced with a large surprise in interest rates.

Our evidence suggests that large firms show little reaction to unexpected changes in monetary policy. However, they may react to expected monetary policy responses to economic conditions. As mentioned before, we cannot measure how business investment reacts to an expected monetary policy response because we don’t know whether, in terms of their investment, firms are reacting to economic conditions or to the monetary policy response to those conditions.

Why are small firms more responsive to unexpected changes in monetary policy?

We look at how unexpected changes in interest rates can influence a firm’s access to credit and thus its investment response. The difference in the investment strategies of small and large firms might be explained by their different access to credit. We start with the idea that it is more expensive for firms to borrow money than it is for them to use internal cash flows. This is true because lenders pass on their costs of evaluating and monitoring borrowers. The premium tends to be larger for newer firms and smaller firms because these firms are generally less established and have a shorter credit history with lenders. This makes it more costly for lenders to monitor and evaluate them.

Bernanke and Gertler (1989) show that the premium firms pay to lenders is not only higher for small firms but also fluctuates more. As a result, small firms are more sensitive to financing conditions when they make investment decisions. One reason for this is that lenders have stricter collateral requirements for small firms to mitigate the risk associated with lending to them. The value of collateral, however, fluctuates with economic and monetary conditions. For example, when interest rates rise unexpectedly, the value of collateral can decrease, thus weakening a firm’s balance sheet. The weaker balance sheet reduces a firm’s ability to borrow. This is particularly true for small firms, which tend to face stricter collateral requirements. Thus, these firms may postpone or forgo some investments.

Unexpected changes in interest rates can also influence the availability of credit in the economy. For example, a higher-than-expected interest rate can increase the riskiness of loans and cause banks to be more cautious and to reduce their loans. When banks are less willing to lend, small firms may feel a greater effect than large firms because they are less able to raise funds through non-bank channels. The lack of external financing means that small firms are more likely to run into difficulties and miss out on investment opportunities.

Conclusion

We find that small firms are more responsive to unexpected monetary policy changes. We reason that this is because their access to credit fluctuates more than that of large firms in response to changing financial conditions. This situation is not unique to Canada. Ehrmann (2000) finds that in Germany the investment response of small firms is stronger than that of large firms, and Cloyne et al. (2018) find that younger firms change their investments more drastically than older firms. These findings suggest two main takeaways:

  • Financial conditions are an important channel in the transmission of monetary policy to economic activity.
  • Firm characteristics can help us better understand fluctuations in business investment.

The scope of this note is limited to firms whose data is available on Compustat. These firms tend to be larger and more established than the private businesses not included in this sample. Additionally, Compustat does not tell us about the age of firms. Cloyne et al. (2018) have shown that age can be an even better proxy for how restricted firms’ access to credit may be. As a result, estimating the investment response of private businesses and incorporating firm age would likely give a clearer picture of the aggregate economic implication of monetary policy surprises.

Appendix

Constructing monetary policy surprises

Following Cloyne et al. (2018) and Gertler and Karadi (2015), we construct a monetary policy surprise shock using a reduced-form vector autoregression (VAR) and futures shocks for identification. The data and variables used in this analysis are described in Table A-1. We do this in three steps outlined below.

Step 1: We estimate the reduced-form system using industrial production (IP), the consumer price index (CPI), unemployment rate, the excess bond premium and yield on one-year Government of Canada bond to obtain estimates of the reduced-form residuals.

Step 2: We regress the residuals of IP, CPI, unemployment and the excess bond premium equation on the residual of the one-year government yield equation using the external instrument BAX surprise. Then we define the residuals of these regressions to be e1, e2, e3 and e4.

Step 3: We regress the residual of the one-year government yield on residuals of the right-hand-side variables, using the e1, e2, e3 and e4 estimated in Step 2 as the instruments. The residual of this regression is defined as the monetary policy surprise.

Table A-1: Data

Variable Source Description
Industrial production (IP) Haver (S156D@G10) Canada: industrial production: manufacturing, mining and utilities (SA, 2007 = 100)
Consumer Price Index (CPI) Haver (S156PC@G10) Canada: consumer price index (SA, 2002 = 100)
Unemployment Haver (S156ELUR@G10) Canada: unemployment rate: 15 years and older (SA, %)
Government bond yield Haver (R156G1@INTDAILY) Canada: treasury bills: 1 year (%)
Excess bond premium (US) Gilchrist’s website Excess bond premium in corporate bonds
3-month BAX futures Bloomberg Banker’s acceptance futures rate
Capex Compustat through WRDS Capital expenditure
Size Compustat through WRDS Total assets
Investment ratio Compustat through WRDS 4 x quarterly capex / previous quarter’s total property, plant and equipment (net)

Estimating the impact of the surprise monetary policy changes on investment

We estimate the investment response to surprise changes in monetary policy following a Jorda-style regression (Jorda 2005) below. This is akin to studying the response of expanding window changes in the investment rate with the monetary policy surprise as the regressor. \(Δ_{h}X_{i.t+h}\) denotes changes in the investment rate from \(t\) to \(t+h\), and \(R_{t}\) denotes the monetary policy surprise at \(t\):

\(Δ_{h}X_{i.t+h} =\, α_i^h +\, β_i^h R_{t} +\, QuarterDummy +\, ε_{i,t+h}\)

Defining small and large firms

At each given quarter, we divide a cross-section of firms into small firms and large firms by comparing the value of their total assets. We label firms whose total assets are greater than the 80th percentile as large, and those whose total assets are less than the 60th percentile as small.

References

  1. Bernanke, B. and M. Gertler. 1989. “Agency Costs, Net Worth, and Business Fluctuations.” The American Economic Review 79 (1): 14–31.
  2. Cloyne J., C. Ferreira, M. Froemel and P. Surico. 2018. “Monetary Policy, Corporate Finance and Investment.” National Bureau of Economic Research Working Paper No. 25366.
  3. Ehrmann, M. 2000, “Firm Size and Monetary Policy Transmission: Evidence from German Business Survey Data.” European Central Bank Working Paper No. 21.
  4. Gertler, M. and P. Karadi. 2015. “Monetary Policy Surprises, Credit Costs, and Economic Activity.” American Economic Journal: Macroeconomics 7 (1): 44–76.
  5. Jordà, O. 2005. “Estimation and Inference of Impulse Responses by Local Projections.” American Economic Review 95 (1): 161–182.
  6. Wilkins, C. A. 2019. “Economic Progress Report: Investing in Growth.” Speech at the Calgary Chamber of Commerce, Calgary, Alberta, May 30.

Avis d’exonération de responsabilité

Les notes analytiques du personnel de la Banque du Canada sont de brefs articles qui portent sur des sujets liés à la situation économique et financière du moment. Rédigées en toute indépendance du Conseil de direction, elles peuvent étayer ou remettre en question les orientations et idées établies. Les opinions exprimées dans le présent document sont celles des auteurs uniquement. Par conséquent, elles ne traduisent pas forcément le point de vue officiel de la Banque du Canada et n’engagent aucunement cette dernière.