Review
- Theories of investment are Keynesian, Neoclassical, and q-theory
- Out of these, q-theory makes the most sense as a "theory of investment"
- The variable q is a "forward looking" variable by including all the information necessary to understand the value of the firm. It's called a sufficient statistic.
- Sommers tests this but finds the coefficients on q to be very small. Does this mean q (ratio of market value to replacement cost) doesn't matter? No. Beta is measured by the covariance of the independent variable and dependent variable divided by the variance on the independent variable. The variance of the stock market is very large, thus, the variance of q should be large, so even if there is some covariance, it will be swamped by the large variance and push down the value of the coefficient artificially.
VAR Results (horserace model)
- Investment responds to: interest rates, output, stock prices, banking conditions, etc.
- This is tested by VAR -- vector autoregression -- which continues to run regressions on sets of variables until a pattern emerges
- Results and more or less consistent with expectations
- However, magnitude of interest rate effect is low
Why does banking matter?
Banking plays a crucial role in economic cycles. There is a long history between financial markets and economic activity. Financial market "frictions" (e.g. information problems) can introduce another mechanism through which money affects output (e.g. upward sloping AS curve). What is the link between economic activity and financial markets? Information.
We will investigate some microeconomic theories to explain the consequences of asymmetric information.
Some Terminology
- Adverse selection: nature of the problem takes place before the contract is signed (e.g. used car markets where you don't have all the information about product quality).
- Moral hazard: problem happens after signing the contract. Examples are insurance.
Stiglitz and Weiss (1981)
Credit rationing appears to exist in the real world. Why? shouldn't demand and supply change until the interest rate reaches an equilibrium? Is the disequilibrium temporary?
The disequilibrium is not temporary. Everything is driven by information problems. According to SW, credit rationing exists because banks don't just care about interest rates but also about loans. The REAL problem is that interest rates affect the risk of the pool of loans because of adverse selection and moral hazard (low rates attract a much larger pool and higher likelihood of risky individuals). It's very difficult for banks to predict who will repay their loans, thus their supply curve has a unique shape. Screening devices (collateral, credit reports, etc.) are costs.
SW Model Particulars
\(\theta\) is an index of the projects
R is the gross returns to the company.
The distribution of returns is given by \(F(R, \theta)\)
Mean preserving spread: same mean, but the variance is higher (see ppt)
Question: When does the default occur? The contractual agreement is as follows:
\(C + R \leq B(1 + \hat r)\)
where B is the loan amount and r is the interest rate. Thus this represents the situation of default, where the return is too little. Unfortunately the entrepreneur cannot guarantee the return they get on their project (R), so they will pay collateral (C).
Net Return to Borrower
The profit off the project is the return to the project, minus the payment (successful project) OR collateral (unsuccessful project):
\(\pi (R, \hat r) = max[R - (1 + \hat r)B, - C]\)
Graphically:
Key Theorems from the Paper:
- For a given interest rate there is a critical value of the projects such that a firm borrows from the bank if and only if their the value of their project is higher.
- As the interest rate increases, the critical value of the projects, below which individuals do not apply for loans, increases.
- The expected return on a loan to a bank is a decreasing function of the riskiness of the loan.
- Bank profits can decrease with an increase in interest rates (critical contribution of paper).
See graphs of select situations below:
Small Shocks, Large Cycles Puzzle
Seemingly small perturbations in the economy can lead to large fluctuations in economic activity. I.e. relatively small changes in interest rates generate large movements in investment spending.
Bernanke, et al. (1996) on the financial accelerator: credit market conditions amplify and propagate the effect of initial real monetary shocks. Key result: internal net worth is very critical for determining finance.
Evidence
First set of tests: augment q-theory style regressions by inclusion of internal measures of liquidity. Exploit the notion that firms may be liquidity constrained. If q-theory is correct, q should be a significant statistic for explaining investment spending (nothing else should matter). The nature of the results is that liquidity constraints do not matter, however, many are still not persuaded by the arguments.
What have we seen so far?
Basic conclusion: financial system affects investment spending in the short run. What about long run? The basic argument is that if a financial system is important for determining the quality of projects undertaken, then there may be long-run consequences as well. The financial structure could enhance growth. However, chicken-egg problem.