Capital Structure - APV
Capital Structure - APV
Capital Structure - APV
Capital Structure
Ernst Maug
University of Mannheim
http://cf.bwl.uni-mannheim.de
maug@corporate-finance-mannheim.de
Tel: +49 (621) 181-1951
Overview
US Germany*)
Type of
Corporate Personal Corporate Personal
income
Dividends Corporate tax Personal income Corporate tax Capped
rate tax, corporate tax rate (15%) witholding tax
not credited (25%)
Retained Corporate tax Capital gains tax Corporate tax Capped
earnings / rate (20%) rate (15%) witholding tax
capital gains (25%)
Interest income Not taxed Personal income Not taxed (except Capped
tax local tax) witholding tax
(25%)
rWACC rOA TC · L· rD
→ With WACC, you consider tax savings from leverage by adjusting the cost of
capital downward.
→ Value company by discounting cash flows to the unlevered firm at rWACC.
→ Investors receive rOA = rWACC + L∙rD∙TC:
- rWACC from the company
- L∙rD∙TC from the government as a tax subsidy for using debt instead of equity
Formula Assumptions
→ Reconsider the perpetuity model from lecture 4 (cash flow example from lecture 2)
→ Assumptions and results:
- FCFU2009: €4.18
- FCFL2009: €4.77
- Nominal growth rate: 4.55%
→ Assumptions about cost of capital:
- rOA 7.00%
- rD 4.00%
- TC 38%
- Target leverage 20%
→ See spreadsheets on website:
- Value Drivers - PerpetuityModel.xls
1.07
Miles-Ezzel (annual) rWACC = 7.00% − 0.38 0.20 4.00% = 6.69%
1.04
U
FCF2009 €4.18
V2008 = = = €195.47
rWACC − G 0.0669 − 0.0455
→ The probability of bankruptcy increases with the debt ratio, hence higher debt
ratios imply higher expected costs of financial distress.
→ The optimum debt ratio trades off increases in the costs of distress against
increases in the tax shield.
Firm Value
PV(Cost of Distress)
PV(Tax shields)
VU
Debt Ratio
Optimum
→ Question: Do firms underlever and choose positions to far too the left of the
optimum?
- Yes: Graham (2000) finds that debt policy is generally conservative
▪ Potential reason: Managers fear for their jobs and are willing to give up some
tax savings (i.e., sacrifice shareholder value to the government)
- No: Molina (2005) finds that expected costs of financial distress are of the same
order of magnitude as the tax benefits from additional debt
→ Conventional procedure:
- multiply
▪ costs of distress (say 20% of firm value)
▪ by probability of bankruptcy (say 5%)
- obtain expected costs of default of 0.2*0.05=0.01 – only 1% of value
- compare with benefits of debt finance (e.g., 3% of value)
0 1 2 3 4 5+
FCFU 100 105 110 116 122 125
Terminal Value 1391
→Step 2: Calculate tax shields and the present value of tax shields (assume: r TXS=rD).
Year 0 1 2 3 4 5+
Debt 1000 800 600 400 200 200
Interest 80 64 48 32 16
Principal repayment 200 200 200 200 0
Debt repayment 280 264 248 232 16
Tax shield 32 26 19 13 6
Terminal value Tax shield 80
PV(Tax shield) 135
→ Assume the tax shield has the same risk as the company itself. Then the APV
formula simplifies:
- rTXS = rOA
- This implies:
FCFtU Int t TC FCFtU + Int t TC
APV = t
+ =
t t
t =1 (1 + rOA ) (1 + rTXS ) t =1 (1 + rOA )
FCFt L
= t
t =1 (1 + rOA )
→ The unlevered cash flow and the tax shield are „compressed.“
→ APV, compressed APV and DCF-WACC will result in different values
- if you make different assumptions about the tax shield.
VL rOA g FCF1U rD TC D
→ Subtract the tax savings from both sides
D
VL rOA − rDTC − g = FCF1U
VL
→ Rearrange
FCF1U FCF1U
VL = =
D
rOA − rDTC − g rWACC − g
VL
→ Obtain WACC-valuation → Yes, this is generally true!
→ Assume that interest payments, dividends, and capital gains are all taxed in the
same way.
- TXS = 0
- OA = E + D
E D
OA = E + D
D +E D +E
D
E = OA + (OA − D )
E
→ See lecture 3.
→ Hamada assumptions:
- Current debt D will be rolled over. It is perpetual and fixed.
- Debt is risk-free, i.e. D = 0
- Tax shield will be fully utilized.
▪ Future earnings will be higher than interest payments.
→ Value of the tax-shield:
- Firm must pay D rF in interest each period.
- This generates a tax saving of TC D rF per period.
- Future tax savings are certain, i.e. TXS = 0.
- The present value of the tax shield is:
TC DrF TC DrF
TXS = = = TC D
t =1 (1 + rF )
t
rF
→ We therefore obtain:
E E
OA = E = E
OA E + D − TC D
D
E = 1 + (1 − TC ) OA
E
→ The company might retire the debt at some point T in the future, so that TSt = 0
for all t > T.
- Then, the present value of the tax-shield is lower than TCD.
→ If firms retire debt in bad times (constant leverage).
- Tax shields TSt are correlated with the market, βTXS > 0.
→ If firms retire debt in good times:
- Pay down very high debt after an LBO
- βTXS will be lower (unlikely that βTXS < 0)
→ This discussion shows that
- Hamada’s assumptions are very special.
- There are good arguments for βTXS > 0.
→ Plug the values TXS = OA and D = 0 into the general unlevering formula
E TXS E E
OA = E − OA OA = E = E
OA OA OA + TXS E +D
D
E = 1 + OA
E
Relever betas → E
Hamada Miles-Ezzel
Yes No
E TXS=0? E
OA = E OA = E
E + D − TC D E +D
Note: when unlevering or relevering, always use the capital structure of the same
firm for which you are calculating the unlevered/relevered beta!
© 2021 Ernst Maug Corporate Finance I 38
DCF Valuation – Listed companies – A roadmap
1. Run regression → E