Advanced Corporate Finance – Lecture 1: Planning and Funding Corporate Investment
Identifying Investment Opportunities
How do firms make investment decisions?
• Common rule of thumb: Invest in any project with NPV>0
• NPV is discounted value of all projected cash flow
𝐶1 𝐶2 𝐶3
−𝐶0 + + 2
+ +. ..
1+𝑟 (1 + 𝑟) (! + 𝑟)3
• Discount rate r is the opportunity cost of capital, the next best return that the firm can obtain elsewhere
o Returns should be adjusted for the difference in risk before comparison
• NPV>0 => Project offers a higher return than the best alternative investment opportunity
Measuring investment opportunities
• Positive NPV means that the value created from the investment exceeds the firm’s initial cost of purchasing capital
• Stock market participants should anticipate this (as they are always forward looking), and they bid up the firm’s share
price even before the investment is undertaken
• Can measure the investment opportunities by comparing firm’s market value to the purchase price of its assets
𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝐹𝑖𝑟𝑚
• 𝑇𝑜𝑏𝑖𝑛’𝑠 𝑄 = 𝑅𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑎𝑝𝑖𝑡𝑎𝑙
o Replacement cost: Price firm would have to pay on the market for capital
o Market-to-book ratio is similar, and sometimes used instead
Introduction to Q Theory
How to interpret Tobin’s Q
• Firms with Q>1 generates more value using capital than other investors or firms would
o Internal value of capital is higher than what the market is willing to pay
o Ex. A Shipping Firm with Q = 1,5 can buy a new truck for 100 and use it to generate a NPV of 50
• Firms with Q<1 is wasting some capital, thus it is better off selling its assets (ex. Badly managed firms)
o The firm should use debt’s market value, but often only the book value is observed
o The denominator of Q is the book value of total assets, or PP&E
→ The book value measures price the firm actually paid to buy the assets on the market
• A firm that has a positive NPV > 0 has an investment opportunity. This can be identified through a Tobin’s Q value > 1
Brief overview of Q theory
• Ex. Firm owns capital 𝑘0 and can spend 𝐼0 on new assets
o Capital depreciates at annual rate 𝛿
𝛼𝐾
• Firm’s value is 𝑉0 = 1+𝑟1, where 𝐾1 = 𝐼0 + (1 − 𝛿)𝐾0
o The firm’s value is the profit that it can generate from its capital next year in the form of its present value; the
capital next year is the capital you currently have adjusted for depreciation and the capital spent on new assets
• Firm also pays capital adjustment costs of (𝐼0 /𝐾0 )^2
o The more capital you create, the higher the attention costs/friction
o Ex. Operating more machines requires increasing amount of attention from management
Firm’s maximization problem (not included in exam)
• Firm chooses 𝐼0 to maximize
𝛼𝐾1 𝐼 2
− 𝐼0 − (𝐾0 )
1+𝑟 0
o The benefits (capital) the firm creates from investment minus the investment, minus the adjustment costs (the
firm looks for the ultimate amount to investment to maximize this equation)
𝑠. 𝑡. 𝐾1 = 𝐼0 + (1 − 𝛿)𝐾0
o This is the constraint here, which is that the capital you have tomorrow, equates to the amount you invest today
plus the amount of capital that did not depreciate
• To solve, set up Lagrangian and take derivative w.r.t. 𝐼0
𝛼𝐾1 𝐼0 2
𝐿= − 𝐼0 − ( ) + 𝑞(𝐼0 + (1 + 𝛿)𝐾0 − 𝐾1 )
1+𝑟 𝐾0
𝜕𝐿 𝐼0 1
= −1 − 2 ( ) ∗ + 𝑞 = 0
𝜕𝐼0 𝐾0 𝐾0
𝐼
𝐾0 = (𝑞 − 1) ∗ 𝐾0 /2
0
o Invest only if q > 1 (as if q < 0, it makes the whole equation negative, suggesting that the best strategy is to not
engage in investments and sell your assets)
• A few technical details
, o In the maximization problem, 𝑞 = 𝜕𝑉1 /𝜕𝐾1. So, investment depends on the marginal profits from one extra unit
of capital
→ Q is measured as the average value of capital (total market value / total capital costs)
→ Hayashi Theorem shows that the average and marginal Q are the same under plausible conditions
o What happens when firm has no adjustment costs?
→ If so, in the maximization problem, q will always equal to 1 (q = 1)
→ Any changes that would lead to q > 1 would cause the firm to invest an infinite amount; since there would be
no net cost to investment
→ Having an adjustment cost is to prevent the firm from committing to an infinite investment as it wouldn’t
make sense
Evidence on Q Theory
• Two formidable hypothesis
o Investments increase with increase in q (firms with higher q invests more than firms with lower q)
o The only variable that matters to investment is q
Empirical evidence on Q theory
• Several problems with Q theory
1. Q explains little of the variation in investment (regression 𝑅2 values are low)
2. Regression estimates of Q’s effect are only consistent with theory if adjustment costs are huge
3. Numerus other variables are significantly related to investment
➔ In summary, findings imply that Q matters for corporate investment, but other factors matter more
Hypothesis 1: Measurement error in Q
• Measurement error could explain the lack of empirical evidence
o In OLD regressions, coefficient is biased towards 0 when the variable contains measurement error, and 𝑅2 falls
as well
o Measurement error means that some of the variables that we measure do not represent the theoretical idea,
leading to biased coefficient, converging towards 0
o Bias can also lead to significant coefficients on other variables
• Q is likely to have some measurement error
o Market value of equity may be influenced by factors other than investment opportunities
o Q should be based on the market value of debt; however, it is often based on the book value of debt
o Book value of assets is based on historical prices, and this may not reflect the current replacement costs
o Most intangible assets have value but are not recorded in the financials
• Erickson and Whited (2000) employ advanced statistical technique to remove measurement error from Q
o Main findings: - 𝑅2 more than doubles after accounting for measurement errors
- Coefficient estimates on Q are substantially larger than the Simple PLS
- Estimates of other variables (e.g. CFs) are small and insignificant
• Peters and Taylor (2017) find that Q works much better when intangible assets are included
Correcting for Q measurement error
• The column “Third” displays a higher 𝑅2 and a higher coefficient for Tobin’s Q, as the measurement errors are
addressed, compared to the standard OLS test
,Main results of Peters and Taylor (2017)
• Premeasured investment by including intangible investment (e.g. R&E) and Peters and Taylor took the Q measure and
added intangible assets to the numerator and denominator
• When intangible investments are added (Panel A), both the 𝑅2 and the coefficient of Q goes up
Hypothesis 2: Financial constraints
• Firms must invest 𝐶0 up front, and receive CFs in the future
• Young firms start with almost no cash, while they are also not profitable
o The main way to fund investment is to raise external financing
o Mature firms typically profitable and they hold a significant amount of cash in hand, but they may face very large
upfront investment costs
• If a firm has difficulty raising external financing, it may pass up investments opportunities even with a higher Q
Why might raising external finances be costly?
• The pecking order theory states that investment is cheaper to finance using internal funds rather than external funds
o This is due to the extra costs associated with raising external funding
• The key reason here is asymmetric information: The firm’s managers will always know its condition better than external
investors
o If managers are trying to reduce ownership stake by selling shares, it might signal that the firm is in bad condition
o Using internal cash signals the firm believes its projects will be profitable
• Implication: Firms with volatile CFs should build up internal funds, to ensure sufficient liquidity during bad economical
years for investments
Example of Precautionary Savings
• Precautionary saving is when firms hold on to its cash during good years in order to ensure enough liquidity in hand to
fund investments, thus avoiding raising external financing due to its costs, during bad economic years
Simple example of precautionary savings
• A firm operates for two years. It has access to two investment opportunities each year.
o Each project 𝑖 = {1,2}requires investment of 𝐼𝑖 and generates return 𝑅𝑖 in the next year (note that the projects
last only one year)
o Suppose 𝐼1 = 𝐼2 = 𝐼, but 𝑅1 > 𝑅2 (suggesting diminishing returns to investment)
• In each year, the firm receives CF 𝐶𝐺 = 2𝐼 with probability 𝜗 (when the economy is doing well), or CF 𝐶𝐵 < 𝐼 with
probability 1 − 𝜗 (when the economy is doing bad)
• In the first year, the firm is in a good state with a good economy, thus receiving 𝐶𝐺
o The firm faces a problem where it must decide whether it should invest all its CF in the given year, or save some
for the next year
o Cash is saved at a RF rate of 0, with no discounting
• Suppose firm invests all CF in Year 1 (no precautionary savings)
o Then in Year 2, it can only invest if the CF is positive
o The Expected Value is:
𝐸[𝑉] = (𝐶𝐺 − 2𝐼 + 𝑅1 + 𝑅2 ) + (𝜗[𝑅1 + 𝑅2 + 𝐶𝐺 − 2𝐼] + (1 − 𝜗)𝐶𝐵 )
𝐸[𝑉] = (1 + 𝜗)(𝑅1 + 𝑅2 ) + (1 − 𝜗)𝐶𝐵
In Year 1, you get CF 𝐶𝐺 , and you spent all the cash on 2 investments (2I); at the end of the year, you receive the
returns from both investments (𝑅1 + 𝑅2 )
, In Year 2, if you are in a good economy with probability 𝜗, you get again CF 𝐶𝐺 and spend it on 2 investments
with returns (𝑅1 + 𝑅2 ) at the end of the year
However, if you are in a bad economy with probability (1 − 𝜗), you get CF 𝐶𝐵 , and no investments will be made
• Instead, if the firm could invest into only one project in Year 1
o It saves 𝐶𝑠 = 𝐶𝐺 − 𝐼 = 𝐼, so in Year 2 it can invest in one project as well, despite the economy state
o Expected Value:
𝐸[𝑉] = (𝐶𝐺 − 𝐶𝑆 − 𝐼 + 𝑅1 ) + (𝜗(𝑅1 + 𝑅2 + 𝐶𝐺 − 2𝐼 + 𝐶𝑆 ) + (1 − 𝜗)(𝑅1 + 𝐶𝑆 − 𝐼 + 𝐶𝐵 )
𝐸[𝑉] = 2𝑅1 + 𝜗(𝑅2 + 𝐶𝑆 ) − (1 − 𝜗)(𝐶𝐵 )
In Year 1, the firm receives CF 𝐶𝐺 , it saves CF 𝐶𝑆 , while investing in only one project, generating return 𝑅1
In Year 2, if the economy is good, the firm will invest in two projects with its CF, receiving return (𝑅1 + 𝑅2 )
However, if the economy is bad, the firm saved enough to invest in one project, receiving return 𝑅1
• Firms are better off to engage in precautionary savings when
2𝑅1 + 𝜗(𝑅2 + 𝐶𝑆 ) + (1 − 𝜗)(𝐶𝐵 ) > (1 + 𝜗)(𝑅1 + 𝑅2 ) + (1 − 𝜗)𝐶𝐵
=> 2𝑅1 + 𝜗(𝑅2 + 𝐼) > 𝑅1 + 𝑅2 + 𝜗𝑅1 + 𝜗𝑅2
=> 𝑅1 − 𝑅2 > 𝜗(𝑅1 − 𝐼)
o This expression shows that firm saves precautionary cash 𝐶𝑆 when:
1. The second project’s return is much lower than the first
→ The value of additional investment in a good economy is less than saving CF
2. Probability of a good economy (𝜗) is low
→ The firm has high likelihood of being constrained in the future
Empirical evidence on financial constraints
• Fazzari, Hubbard, and Petersen (1988) provided early evidence that financial constraints affect investment
• Split firms into 3 groups based on dividend payouts
o Firms that pay dividends instead of saving cash are likely to be not financial constrained
o Tested relationships between annual CFs and investment across each subset of firms
• Data shows that the effect of CFs is the strongest for most constrained firms, and the weakest among the least
constrained firms
o This indicates that investment depends on the availability of internal funds
• Main result
Critiques from Kaplan and Zingales (1997)
• Empirically refuted Fazzari, Hubbard, and Petersen’s results, showing that financial constraints may not affect
investment-CF sensitivity
• Re-classified firms based on the discussion about constraints in 10-K filings
o Many “constrained” firms mentioned no problems with raising funds
• They found that CFs are associated with investment across all firms, regardless of financial constraints
o The association is the strongest for the least constrained firms
• Additional problem with Fazzari, Hubbard, and Petersen’s results is that: CFs can be correlated with investment
opportunities, thus causing biased regression estimates
Constraints and precautionary savings