Finance Cheat Sheet

Finance Cheat Sheet

Part of: Financial Management Reading 1 — Financial Management & Finance Source: Finance Cheat Sheet.pdf (4 pages, Amitai Malik, Teaching Success)

This is a summary sheet, not a taught lesson

It is the course formula sheet: every formula the course uses on four pages, with a few worked examples. It sits under Readings on the site. Use it to revise and look up formulas; to learn a topic, follow the link in each section to the lesson that teaches it. The sheet has two blank "Personal Additions" boxes, which are not reproduced.

How the sheet is organised

1. Time value of money

Taught in Lesson_01-Time Value of Money and Lesson_02-Perpetuities, NPV & IRR. Practice: HW_01-Time Value of Money.

rr is the discount rate and, unless the question says otherwise, it is compounding.

  • Discounting = moving to the left on the timeline: divide by (1+r)n(1+r)^n.
  • Compounding = moving to the right: multiply by (1+r)n(1+r)^n.

Here nn is the number of periods (or payments, for annuities), PMTPMT the periodic payment, gg the periodic growth rate.

Cash flow Present value Future value at nn
Single cash flow FVn(1+r)n\dfrac{FV_n}{(1+r)^n} PV (1+r)nPV\,(1+r)^n
Annuity (finite, identical payments) PMTr[1−1(1+r)n]\dfrac{PMT}{r}\left[1-\dfrac{1}{(1+r)^n}\right] PMTr[(1+r)n−1]\dfrac{PMT}{r}\left[(1+r)^n-1\right]
Growing annuity PMT1r−g[1−(1+g1+r)n]\dfrac{PMT_1}{r-g}\left[1-\left(\dfrac{1+g}{1+r}\right)^n\right] PMT1r−g[(1+r)n−(1+g)n]\dfrac{PMT_1}{r-g}\left[(1+r)^n-(1+g)^n\right]
Perpetuity (infinite, identical payments) PMTr\dfrac{PMT}{r} none
Growing perpetuity PMT1r−g\dfrac{PMT_1}{r-g} none
Two timing rules from the sheet
  • PV takes you one period before the first payment.
  • FV takes you to the final payment.

Plain English. The PV of an annuity is "payment divided by the rate, minus the part that would have been paid after year nn". A growing perpetuity is the same idea with the rate reduced by the growth.

Special cases and tools:

g=r:PV=n⋅PMT11+r,FV=n⋅PMT1 (1+r)n−1g = r:\qquad PV=\frac{n\cdot PMT_1}{1+r},\qquad FV=n\cdot PMT_1\,(1+r)^{n-1}

When growth equals the discount rate, every payment has the same present value, so the PV is just nn equal pieces.

PMTk=PMT1 (1+g)k−1PMT_k = PMT_1\,(1+g)^{k-1}

The kk-th payment of a growing stream.

g≥r  ⇒  PV→∞(growing perpetuity)g \ge r \;\Rightarrow\; PV\to\infty \quad\text{(growing perpetuity)}

A growing perpetuity only has a finite price if r>gr>g.

Extracting the power (solving for nn): if 1.1n=1.46411.1^n=1.4641, then

n=ln⁡1.4641ln⁡1.1=4n=\frac{\ln 1.4641}{\ln 1.1}=4

Take logs of both sides to bring nn down from the exponent.

Excel: PV, FV, PMT, NPER and RATE cover the annuity rows; for a stream of different amounts use NPV.


2. Interest rates: stated vs effective

Taught in Lesson_03-Rate Conversion, Effective Rates & EAC. Practice: HW_04-Effective Interest Rates.

Three ground rules from the sheet: always draw a timeline; adjust every rate to the same period as the timeline (monthly payments need a monthly rate); and when comparing scenarios, find each one's PV and compare.

nn = how many times the shorter period fits into the longer one.

To the shorter period To the longer period
Stated interest (APR) rlongern\dfrac{r_{longer}}{n} n⋅rshortern\cdot r_{shorter}
Effective interest (EAR) (1+rlonger)1n−1(1+r_{longer})^{\frac{1}{n}}-1 (1+rshorter)n−1(1+r_{shorter})^{n}-1

Stated rates are split or multiplied; effective rates are compounded or rooted. See Stated Rate and Effective Interest Rate.

When stated and effective differ

The sheet lists three situations.

1. Interest compounded more than once a period (nn = number of compoundings):

re=(1+rsn)n−1r_e=\left(1+\frac{r_s}{n}\right)^{n}-1

Example from the sheet: APR =12%=12\% compounded monthly gives

EAR=(1+0.1212)12−1=12.68%EAR=\left(1+\frac{0.12}{12}\right)^{12}-1=12.68\%

2. Interest paid in advance (nn = the transaction's length):

re=(11−rs)1n−1r_e=\left(\frac{1}{1-r_s}\right)^{\frac{1}{n}}-1

You receive less than you borrow, so the true rate is higher than the stated one. Example: APR =12%=12\%, one year, paid in advance:

re=(11−0.12)11−1≈13.64%r_e=\left(\frac{1}{1-0.12}\right)^{\frac{1}{1}}-1\approx 13.64\%
Sheet rounding

The sheet prints 13.63%13.63\%. The exact value is 1/0.88−1=0.13636…1/0.88-1=0.13636\ldots, which rounds to 13.64%13.64\%. The sheet truncated it.

3. Commissions. Draw the timeline, then solve the equation below for rer_e, discounting with rer_e itself:

PVincoming=PVoutgoingPV_{incoming}=PV_{outgoing}

Example from the sheet: a $100 loan for one year at APR =12%=12\% plus a $4 commission paid at expiry. At time 1 you pay back 112+4=116112+4=116:

100=1161+re  ⇒  re=16%100=\frac{116}{1+r_e}\;\Rightarrow\; r_e=16\%

3. Inflation and the Fisher equation

Taught in Lesson_04-Inflation, Real Rates & Expected Inflation. Practice: HW_05-Inflation & Real Annuities.

infinf is periodic inflation. Actual inflation from time 0 to nn, from a price index or from period inflations:

inf0→n=Final IndexBase Index−1orinf0→n=(1+inf1)(1+inf2)⋯(1+infn)−1inf_{0\to n}=\frac{\text{Final Index}}{\text{Base Index}}-1 \qquad\text{or}\qquad inf_{0\to n}=(1+inf_1)(1+inf_2)\cdots(1+inf_n)-1

Cumulative inflation is a ratio of price levels, or a product of the yearly growth factors. Never a sum.

Sheet examples:

  • Base index M0=100M_0=100, M2=120M_2=120: inf0→2=120100−1=20%inf_{0\to2}=\frac{120}{100}-1=20\%. If M2=80M_2=80: inf0→2=−20%inf_{0\to2}=-20\%.
  • Yearly inflations 1%,2%,3%1\%,2\%,3\%: (1.01)(1.02)(1.03)−1=6.11%(1.01)(1.02)(1.03)-1=6.11\%.

Index-linked vs non-index-linked

Real value (without inflation) Nominal value (money in our pocket)
Index-linked (real rate; compensates for time and risk only) FVn=PV(1+rreal)nFV_n=PV(1+r_{real})^n FVn=PV(1+rreal)n (1+inf0→n)FV_n=PV(1+r_{real})^n\,(1+inf_{0\to n})
Non-index-linked (nominal rate; compensates for time, risk and expected inflation) FVn=PV(1+rnominal)n1+inf0→nFV_n=\dfrac{PV(1+r_{nominal})^n}{1+inf_{0\to n}} FVn=PV(1+rnominal)nFV_n=PV(1+r_{nominal})^n

Which is better, in the sheet's words:

  1. If inf0→n<πexpinf_{0\to n}<\pi_{exp} (actual below expected), non-index-linked is better. If inf0→n>infexpinf_{0\to n}>inf_{exp}, index-linked is better.
  2. Non-index-linked is more dangerous: we cannot know whether we will withstand future payments.

Fisher equation

(1+rnominal)=(1+rreal)⋅(1+infexp)(1+r_{nominal})=(1+r_{real})\cdot(1+inf_{exp})

Nominal growth = real growth times expected price growth. See Fisher Equation. Two conditions: all variables must cover the same time period, and all must be effective (not stated) rates.

Worked example from the sheet. A bank offers a quarterly effective 3% on index-linked deals and a semi-annual effective 10% on non-index-linked deals. What annual inflation does the bank expect? First put both on the annual period:

rreal, annual=(1.03)4−1=12.55%rnominal, annual=(1.1)2−1=21%r_{real,\,annual}=(1.03)^4-1=12.55\% \qquad r_{nominal,\,annual}=(1.1)^2-1=21\%

Then plug into Fisher:

1.21=1.1255⋅(1+infexp)  ⇒  infexp=7.5%1.21=1.1255\cdot(1+inf_{exp})\;\Rightarrow\; inf_{exp}=7.5\%

4. Bonds

Taught in Lesson_07-Bond Pricing. Practice: HW_08-Bond Pricing.

Notation: ii = coupon rate (stated), r=YTMr=YTM = market rate (effective), FVFV = face value, PVPV = the bond's price, mm = coupons per year, nn = number of coupons.

First step in every bond question: find the coupon. See Coupon and Face Value.

C=FV⋅imC=FV\cdot\frac{i}{m}

Classic bond (equal coupons, FVFV at maturity), price at issuance:

PV=Cr[1−1(1+r)n]+FV(1+r)nPV=\frac{C}{r}\left[1-\frac{1}{(1+r)^n}\right]+\frac{FV}{(1+r)^n}

An annuity of coupons plus the discounted face value.

Zero-coupon bond (Zero-Coupon Bond), and coupons in perpetuity (no FVFV):

PV=FV(1+r)nPV=CrPV=\frac{FV}{(1+r)^n} \qquad\qquad PV=\frac{C}{r}

Price after issuance, kk coupons already paid:

PVk=Cr[1−1(1+r)n−k]+FV(1+r)n−k(immediately after coupon k)PV_k=\frac{C}{r}\left[1-\frac{1}{(1+r)^{n-k}}\right]+\frac{FV}{(1+r)^{n-k}} \quad\text{(immediately after coupon } k\text{)}
PVk=Cr[1−1(1+r)n−k]+FV(1+r)n−k+C(right before coupon k is paid)PV_k=\frac{C}{r}\left[1-\frac{1}{(1+r)^{n-k}}\right]+\frac{FV}{(1+r)^{n-k}}+C \quad\text{(right before coupon } k\text{ is paid)}

Just before a coupon is handed out, you still get it, so add one more CC.

Discount, par, premium (see Yield to Maturity):

Issued at Condition
Discount YTM>iYTM>i, so FV>PVFV>PV
Par YTM=iYTM=i, so FV=PVFV=PV
Premium YTM<iYTM<i, so FV<PVFV<PV

Credit rating. Safer bond, higher rating, lower market rate, higher price. Dangerous bond, lower rating, higher market rate, lower price.

Worked example from the sheet. 3-year bond, annual 10% coupons, FV=50,000FV=50{,}000. 30% of FVFV is repaid at the end of year 1, the rest at maturity. r=8%r=8\%. Cash flows (this is not a classic bond, so discount each flow):

Time Coupon Principal repaid Total
1 50,000⋅0.1=5,00050{,}000\cdot0.1=5{,}000 0.3⋅50,000=15,0000.3\cdot50{,}000=15{,}000 20,00020{,}000
2 0.7⋅50,000⋅0.1=3,5000.7\cdot50{,}000\cdot0.1=3{,}500 none 3,5003{,}500
3 0.7⋅50,000⋅0.1=3,5000.7\cdot50{,}000\cdot0.1=3{,}500 0.7⋅50,000=35,0000.7\cdot50{,}000=35{,}000 38,50038{,}500
PV=20,0001.08+3,5001.082+38,5001.083=52,081.75PV=\frac{20{,}000}{1.08}+\frac{3{,}500}{1.08^2}+\frac{38{,}500}{1.08^3}=52{,}081.75

Excel: PV prices a classic bond; for irregular cash flows discount each flow with NPV; RATE or IRR recovers the YTM from a price.


5. Loans

Taught in Lesson_05-The Four Loan Types. Practice: HW_06-Loans & Amortization.

Notation: PVkPV_k = starting balance at time kk, rr = interest rate, nn = number of payments, and for period kk: PRNkPRN_k = principal part, INTkINT_k = interest part, PMTkPMT_k = payment.

Annuity loan (Spitzer loan): constant payment

First step: find the payment from

PVn=PMTr[1−1(1+r)n]PV_n=\frac{PMT}{r}\left[1-\frac{1}{(1+r)^n}\right]

Amortization row for period kk:

Starting balance Payment Interest Principal Ending balance
PVkPV_k PMTPMT PVk⋅rPV_k\cdot r PMT−INTkPMT-INT_k PVk+1=PVk−PRNk=PMTr[1−1(1+r)n−k]PV_{k+1}=PV_k-PRN_k=\dfrac{PMT}{r}\left[1-\dfrac{1}{(1+r)^{n-k}}\right]

Totals and a useful pattern:

∑k=1nPMTk=n⋅PMT,∑k=1nINTk=n⋅PMT−PV1,∑k=1nPRNk=PV1\sum_{k=1}^{n}PMT_k=n\cdot PMT,\qquad \sum_{k=1}^{n}INT_k=n\cdot PMT-PV_1,\qquad \sum_{k=1}^{n}PRN_k=PV_1
PRNk=PRN1 (1+r)k−1PRN_k=PRN_1\,(1+r)^{k-1}

In a Spitzer loan the principal repaid grows by a factor (1+r)(1+r) each period.

Regular loan: constant principal

First step: the periodic principal.

PRN=PVnPRN=\frac{PV}{n}
Starting balance Principal Interest Payment Ending balance
PVkPV_k PRNPRN PVk⋅rPV_k\cdot r PRN+PVk⋅rPRN+PV_k\cdot r PVk+1=PV1−k⋅PRNPV_{k+1}=PV_1-k\cdot PRN
∑k=1nPMTk=PV1[1+r(n+1)2],∑k=1nINTk=PV1⋅r(n+1)2,∑k=1nPRNk=PV1\sum_{k=1}^{n}PMT_k=PV_1\left[1+\frac{r(n+1)}{2}\right],\qquad \sum_{k=1}^{n}INT_k=PV_1\cdot\frac{r(n+1)}{2},\qquad \sum_{k=1}^{n}PRN_k=PV_1

Payments shrink each period because interest falls with the balance.

Bullet loan and balloon loan

Bullet (Bullet Loan): one single payment at the end.

PMT=PV(1+r)n,INT=PV[(1+r)n−1],PRN=PVPMT=PV(1+r)^n,\qquad INT=PV\left[(1+r)^n-1\right],\qquad PRN=PV

Balloon: periodic interest-only payments, then the whole principal at the end.

INTk=PV⋅r,PRNn=PVINT_k=PV\cdot r,\qquad PRN_n=PV
Index-linked vs non-index-linked loans
  1. For an index-linked loan, multiply the result by actual inflation (1+inf0→n)(1+inf_{0\to n}) to get the nominal value.
  2. A non-index-linked loan already contains the nominal rate (with expected inflation inside), so multiply by nothing at the end.

Excel: PMT, IPMT, PPMT, PV, RATE.


6. Project choice: NPV, IRR and friends

Taught in Lesson_02-Perpetuities, NPV & IRR (NPV, IRR), Lesson_03-Rate Conversion, Effective Rates & EAC (profitability index, EAC) and Lesson_06-Capital Budgeting & Free Cash Flow. Practice: HW_02-NPV, IRR & Project Choice, HW_03-Profitability Index & EAC.

The two defining formulas:

NPV=−I0+Cf11+r+Cf2(1+r)2+⋯+Cfn(1+r)nNPV=-I_0+\frac{Cf_1}{1+r}+\frac{Cf_2}{(1+r)^2}+\cdots+\frac{Cf_n}{(1+r)^n}
0=−I0+Cf11+IRR+Cf2(1+IRR)2+⋯+Cfn(1+IRR)n0=-I_0+\frac{Cf_1}{1+IRR}+\frac{Cf_2}{(1+IRR)^2}+\cdots+\frac{Cf_n}{(1+IRR)^n}

Net Present Value is today's value of all the cash flows; the Internal Rate of Return is the discount rate that makes it zero. Excel: NPV (remember it starts at period 1, so add −I0-I_0 separately), IRR.

The decision tree on the sheet

Situation Rule
Single project, NPV test NPV>0NPV>0: accept. NPV=0NPV=0: indifferent. NPV<0NPV<0: reject.
Single project, IRR test (only if the project is standard) IRR>rIRR>r: accept. IRR=rIRR=r: indifferent. IRR<rIRR<r: reject.
Multiple, mutually exclusive Pick the greatest NPV, or the greatest MIRRMIRR (below).
Multiple, not mutually exclusive, unlimited budget Undertake every project with NPV>0NPV>0.
Multiple, not mutually exclusive, budget constraint, partial projects not allowed Choose the combination with the greatest cumulative NPV.
Multiple, not mutually exclusive, budget constraint, partial projects allowed Rank by profitability index P.I.=NPV∣I0∣P.I.=\dfrac{NPV}{\lvert I_0\rvert}.
Renewable projects with unequal lives, or continuous projects Pick the greatest EAVEAV (below).
When the IRR test is safe

The sheet says to use the IRR rule only if the project is "standard" (one sign change in the cash flows: outflow first, then inflows). For non-standard projects, trust NPV.

Modified IRR (Modified Internal Rate of Return):

MIRR=(FVall positive cash flowsPVall negative cash flows)1n−1MIRR=\left(\frac{FV_{\text{all positive cash flows}}}{PV_{\text{all negative cash flows}}}\right)^{\frac{1}{n}}-1

Compound all inflows forward to the end, discount all outflows back to today, and take the single-period growth rate between them.

Equivalent annual value (Equivalent Annual Cost is the cost-side version), for projects with unequal lives:

EAV=NPV⋅r[1−1(1+r)n]EAV=\frac{NPV\cdot r}{\left[1-\dfrac{1}{(1+r)^n}\right]}

Turn each project's NPV into a level yearly amount over its own life, so projects of different lengths become comparable.


7. Risk and reward

No lesson in this course covers this section

The course has no lesson on risk, return, portfolios, beta or the SML. The only note that practises it is HW_09-Risk, Return & Diversification, so treat this sheet as the main source for the formulas and the homework as the worked practice. Plain-English glosses below are mine; the formulas are the sheet's.

Learning order that makes this section easy: (a) one asset, (b) two or three assets, (c) add a risk-free asset, (d) market portfolio and the CML, (e) beta and the SML, (f) split total risk into market and specific.

7.1 One asset

Based on scenarios, where PiP_i is the scenario's probability and RiR_i the asset's return in that scenario:

E(R)=∑Pi RiVAR(R)=∑Pi (Ri−E(R))2E(R)=\sum P_i\,R_i \qquad VAR(R)=\sum P_i\,\bigl(R_i-E(R)\bigr)^2

Expected return is the probability-weighted average; variance is the probability-weighted squared distance from it. See Expected Return.

Based on historical data (nn observations):

E(Ri)=1n∑i=1nRi=R‾VAR(Ri)=∑i=1n(Ri−E(Ri))2n−1E(R_i)=\frac{1}{n}\sum_{i=1}^{n}R_i=\overline{R} \qquad VAR(R_i)=\frac{\sum_{i=1}^{n}\bigl(R_i-E(R_i)\bigr)^2}{n-1}

With history, use the plain average and divide the variance by n−1n-1 (not nn).

7.2 Two assets

XA,XBX_A,X_B are the investment proportions and must add to 1:

XA+XB=1,XA=1−XB=Initial sum invested in AInitial investment totalX_A+X_B=1,\qquad X_A=1-X_B=\frac{\text{Initial sum invested in }A}{\text{Initial investment total}}

Correlation measures how the two returns move together (Correlation):

ρAB=Cov(A,B)σA⋅σB,−1≤ρ≤1\rho_{AB}=\frac{Cov(A,B)}{\sigma_A\cdot\sigma_B},\qquad -1\le\rho\le1
Covariance Correlation
Cov(A,B)>0Cov(A,B)>0 ρ>0\rho>0 (move together)
Cov(A,B)=0Cov(A,B)=0 ρ=0\rho=0 (unrelated)
Cov(A,B)<0Cov(A,B)<0 ρ<0\rho<0 (move opposite)

Portfolio expected return and variance (Portfolio Risk):

E(Rp)=XA E(RA)+XB E(RB)E(R_p)=X_A\,E(R_A)+X_B\,E(R_B)
VAR(Rp)=σp2=XA2σA2+XB2σB2+2XAXB ρAB σAσBVAR(R_p)=\sigma_p^2=X_A^2\sigma_A^2+X_B^2\sigma_B^2+2X_AX_B\,\rho_{AB}\,\sigma_A\sigma_B

The return is a plain weighted average, but the risk is not: the cross term shrinks as ρ\rho falls, which is the whole point of Diversification.

For three assets (AA, BB, CC):

E(Rp)=∑i=ACXi E(Ri)E(R_p)=\sum_{i=A}^{C}X_i\,E(R_i)
σp2=∑i=ACXi2σi2+2(XAXBρABσAσB+XAXCρACσAσC+XBXCρBCσBσC)\sigma_p^2=\sum_{i=A}^{C}X_i^2\sigma_i^2+2\left(X_AX_B\rho_{AB}\sigma_A\sigma_B+X_AX_C\rho_{AC}\sigma_A\sigma_C+X_BX_C\rho_{BC}\sigma_B\sigma_C\right)

Every asset's own variance, plus twice every pair's covariance term.

7.3 Minimum variance portfolio (MVP)

The mix of AA and BB with the lowest possible risk (Minimum Variance Portfolio). The sheet gives the general formula for ρ≠±1,0\rho\ne\pm1,0 (it also reproduces the two shortcuts below):

XA=σB2−ρAB σAσBσA2+σB2−2ρAB σAσB,XB=1−XAX_A=\frac{\sigma_B^2-\rho_{AB}\,\sigma_A\sigma_B}{\sigma_A^2+\sigma_B^2-2\rho_{AB}\,\sigma_A\sigma_B},\qquad X_B=1-X_A

Shortcuts for the three standard cases:

Case MVP weight What the curve looks like
ρ=−1\rho=-1 XA=σBσA+σBX_A=\dfrac{\sigma_B}{\sigma_A+\sigma_B} Two straight lines meeting on the vertical axis: risk can be driven to zero
ρ=0\rho=0 XA=σB2σA2+σB2X_A=\dfrac{\sigma_B^2}{\sigma_A^2+\sigma_B^2} A curve bending back to the left of AA
ρ=1\rho=1 Xless risky asset=1X_{\text{less risky asset}}=1 A straight line from AA to BB: no diversification, so the MVP is simply the less risky asset

Read01_two_asset_frontier

About this figure

Theory diagram with illustrative numbers I chose (σA=10%\sigma_A=10\%, σB=20%\sigma_B=20\%, E(RA)=6%E(R_A)=6\%, E(RB)=10%E(R_B)=10\%), not real data. It redraws the three small sketches on the sheet. The dots mark the MVP for ρ=−1\rho=-1 (zero risk) and ρ=0\rho=0.

7.4 Adding a risk-free asset

With one risky asset AA and a risk-free asset rFr_F:

E(Rp)=XA E(RA)+XrF rFVAR(Rp)=XA2σA2σp=XA σAE(R_p)=X_A\,E(R_A)+X_{r_F}\,r_F \qquad VAR(R_p)=X_A^2\sigma_A^2 \qquad \sigma_p=X_A\,\sigma_A

The risk-free part has no variance and no covariance, so portfolio risk is just the risky share times the risky asset's risk.

Leveraged portfolio. The investor borrows at the risk-free rate to invest more than their own money, so XrF<0X_{r_F}<0:

XrF=−Borrowed sumInitial investment total,XA=Initial sum invested in AInitial investment totalX_{r_F}=-\frac{\text{Borrowed sum}}{\text{Initial investment total}},\qquad X_A=\frac{\text{Initial sum invested in }A}{\text{Initial investment total}}

The same formulas for E(Rp)E(R_p) and σp\sigma_p apply. Leverage raises expected return in exchange for higher risk.

Short selling is another way to lever up: sell one stock short to buy more of another.

Xpurchased=Initial sum invested in purchased assetInitial investment total,Xshort sold=−Sum asset was short sold forInitial investment totalX_{\text{purchased}}=\frac{\text{Initial sum invested in purchased asset}}{\text{Initial investment total}},\qquad X_{\text{short sold}}=-\frac{\text{Sum asset was short sold for}}{\text{Initial investment total}}

Check: the weights still add to 1, the short-sold asset's weight is negative.

7.5 Efficient portfolio and the CML

An efficient portfolio combines the risk-free asset rFr_F with the market portfolio MM (all risky assets in the market, held in proportion to their market worth).

  • Step 1 is the same for every investor: determine MM.
  • Step 2 is personal: choose how to split between MM and rFr_F (the more risk averse, the closer to rFr_F).
Xasset=Asset’s market worthTotal worth of the market’s portfolio,Asset’s market worth=no. of shares⋅price per shareX_{asset}=\frac{\text{Asset's market worth}}{\text{Total worth of the market's portfolio}},\qquad \text{Asset's market worth}=\text{no. of shares}\cdot\text{price per share}

Where an efficient portfolio sits:

Position Weights
On rFr_F XrF=1,  XM=0X_{r_F}=1,\;X_M=0
Between rFr_F and MM 0<XrF,XM<10<X_{r_F},X_M<1
On MM XrF=0,  XM=1X_{r_F}=0,\;X_M=1
Right of MM (leveraged) XrF<0,  XM>1X_{r_F}<0,\;X_M>1
E(Rp)=XrF rF+XM E(RM)σp=XM σME(R_p)=X_{r_F}\,r_F+X_M\,E(R_M) \qquad \sigma_p=X_M\,\sigma_M

The Capital Market Line (Capital Market Line) links an efficient portfolio's expected return to its risk. Always leave E(Rp)E(R_p) and σp\sigma_p as variables:

CML:E(Rp)=rf+(E(Rm)−rfσm)⋅σpCML:\qquad E(R_p)=r_f+\left(\frac{E(R_m)-r_f}{\sigma_m}\right)\cdot\sigma_p

Start at the risk-free rate and add the market's extra return per unit of risk, times the risk you take.

Read01_cml_frontier

About this figure

Theory diagram, schematic and not to scale; the shapes are mine, redrawing the sheet's page-4 sketch. The CML touches the efficient frontier of risky assets at MM. AA, BB, CC are single assets that sit below the CML.

7.6 Inefficient portfolios, beta and the SML

A portfolio that is not a mix of the risk-free asset and the market portfolio is inefficient. Single assets and inefficient portfolios are priced by the Security Market Line (Security Market Line), which fits all portfolios and assets:

SML:E(R)=rf+[E(Rm)−rf]⋅βSML:\qquad E(R)=r_f+\bigl[E(R_m)-r_f\bigr]\cdot\beta

Required return = risk-free rate + market risk premium times the asset's sensitivity to the market.

  • The risk premium is E(Rm)−rfE(R_m)-r_f.
  • If you are given β\beta, always use the SML, even for an efficient portfolio.
  • The E(R)E(R) from the SML can be used as the discount rate: E(R)=rE(R)=r.

Beta (Beta) is how sensitive an asset is to market moves.

  • Rule 1: the greater the β\beta, the more sensitive the asset, the higher the return investors demand.
  • Rule 2: the market portfolio has βM=1\beta_M=1, and a risk-free asset has βrF=0\beta_{r_F}=0.
Single asset:β=ρA,M⋅σAσm=Cov(A,M)σm2\text{Single asset:}\quad \beta=\frac{\rho_{A,M}\cdot\sigma_A}{\sigma_m}=\frac{Cov(A,M)}{\sigma_m^2}
Any portfolio:βp=∑i=1nXi βiEfficient portfolio:β=XM\text{Any portfolio:}\quad \beta_p=\sum_{i=1}^{n}X_i\,\beta_i \qquad \text{Efficient portfolio:}\quad \beta=X_M

A portfolio's beta is the weighted average of its parts' betas, and an efficient portfolio's beta is just its share in the market portfolio, because the risk-free part has beta 0.

Regression form (rA,rf,rmr_A,r_f,r_m are the actual returns of asset AA, the risk-free asset and the market):

rA−rf=β (rM−rf)+ϵAr_A-r_f=\beta\,(r_M-r_f)+\epsilon_A

ϵA\epsilon_A is the part of AA's return that is not related to the market.

7.7 Market risk vs specific risk

Total variance splits into the part driven by the market (systematic, undiversifiable) and the part unique to the asset (specific, also called diversifiable, unsystematic or idiosyncratic). See Systematic Risk and Idiosyncratic Risk.

σtotal2=σsystematic2+σspecific2\sigma^2_{total}=\sigma^2_{systematic}+\sigma^2_{specific}
σsystematic2=βa2⋅σM2σspecific2=σtotal2−σsystematic2\sigma^2_{systematic}=\beta_a^2\cdot\sigma_M^2 \qquad \sigma^2_{specific}=\sigma^2_{total}-\sigma^2_{systematic}
% of systematic risk=Systematic RiskTotal Risk=σsystematic2σtotal2\%\text{ of systematic risk}=\frac{\text{Systematic Risk}}{\text{Total Risk}}=\frac{\sigma^2_{systematic}}{\sigma^2_{total}}

7.8 Reading the CML / efficient-frontier diagram

The sheet's last diagram (page 4) puts risky assets A,B,CA,B,C around the efficient frontier. Its bullet points:

  • MM is where Xm=1X_m=1 and Xrf=0X_{r_f}=0: the intersection of the CML and the efficient frontier of all risky assets.
  • The market portfolio contains all risky assets in proportion to their market worth. AA, BB, CC are points on the SML.
  • σB\sigma_B (total risk of BB) = market risk (undiversifiable) up to the CML, plus specific risk (diversifiable) from the CML to BB.
  • The closer an asset is to the CML, the better its correlation with the market. The gap between the asset and the CML is ϵ\epsilon (epsilon).

Comparing assets AA and CC (the sheet's example):

Comparison Answer
Identical total risk (σA=σC\sigma_A=\sigma_C) Same distance from the vertical axis
Greater market risk AA (further from the axis, up to the CML)
Greater specific risk CC (further from the CML)
Greater correlation with the market AA (ϵA<ϵC\epsilon_A<\epsilon_C)

8. Capital budgeting: free cash flow

Taught in Lesson_06-Capital Budgeting & Free Cash Flow. Practice: HW_07-Free Cash Flow & Project NPV.

Steps to solve a capital budgeting question: (1) analyse the information; (2) build a final table with all the project's variables; (3) find the NPV to judge profitability.

Straight-line depreciation (Depreciation):

DEP=I0−SVGNN=Initial Investment−Salvage ValueNo. of depreciation yearsDEP=\frac{I_0-SVG_N}{N}=\frac{\text{Initial Investment}-\text{Salvage Value}}{\text{No. of depreciation years}}

The free cash flow table

Rows, from top to bottom (Free Cash Flow), one column per time 0,1,…,n0,1,\ldots,n:

Row
+ Revenues
− Expenses
− Depreciation
= Operational Profit (EBIT)
− Corporate tax expenses
= Net Operating Profit after Tax (NOPAT)
+ Depreciation (add back, it is not a cash cost)
+ Tax-free revenues
− Undeductible expenses
− Cost of project or initial investment
+ Money received for the sale of fixed assets
− Capital gains tax for the sale of fixed assets
− Changes in working capital
= Free Cash Flow

Rules the sheet attaches to the table:

  1. Any revenue or expense is taxable unless noted otherwise.
  2. If there are several revenues or expenses, each goes on a separate row.
  3. Sometimes irrelevant information is given and must be ignored:
    • A. Sunk costs: an expense that has already happened or was already promised in the future.
    • B. Funding (loans, bonds) and financing expenses.
    • C. Payroll: if the existing workforce is used with no addition to their pay, ignore their wages; if there is an increase in their wages due to the project, count it as a project cost.
  4. General rule when information is misleading: what happens with the project minus what happens without the project is what goes in the table.

Sale of fixed assets

Sometimes equipment is sold before it is fully depreciated. The accounting (book) value differs from the sale price, and the gain is taxed.

Book Value=I0−n⋅DEPCapital Gain=Selling Value−Book Value\text{Book Value}=I_0-n\cdot DEP \qquad \text{Capital Gain}=\text{Selling Value}-\text{Book Value}
Capital Gain Tax=Capital Gain⋅Capital Gains Tax Rate\text{Capital Gain Tax}=\text{Capital Gain}\cdot\text{Capital Gains Tax Rate}

Worked example from the sheet. A mining company buys tractors for $500,000 for a project, depreciated over 5 years with no salvage value. After 3 years they are sold for $300,000. Capital gains tax is 20%.

DEP=500k−05=100kBook Value=500k−3(100k)=200kDEP=\frac{500k-0}{5}=100k \qquad \text{Book Value}=500k-3(100k)=200k
Capital Gain=300k−200k=100kCapital Gain Tax=100k⋅0.2=20k\text{Capital Gain}=300k-200k=100k \qquad \text{Capital Gain Tax}=100k\cdot0.2=20k

In the table: +$300,000 (sale proceeds) and −$20,000 (capital gains tax). The company is left with $280,000.

Investment in working capital

Working capital is the cash tied up in running the business day to day:

Working Capital=Inventory+Customers−Suppliers\text{Working Capital}=\text{Inventory}+\text{Customers}-\text{Suppliers}

Inventory is stock not yet sold, customers is what customers still owe you, suppliers is what you still owe. See Working Capital.

The change in working capital is the row that goes in the free cash flow table (it is "part of the original table", the sheet stresses). An increase in working capital is a cash outflow; a decrease is an inflow.

Worked example from the sheet. "ABC Computers" just started. A customer buys a computer and pays $6,000 in 6 monthly installments; ABC buys the same computer from "HM" for $6,000 cash. So the owners must put in $6,000 of working capital at time 0, and take $1,000 out each month as the customer pays.

Time 0 1 2 3 4 5 6
Inventory 0 0 0 0 0 0 0
Customers (still owed) 6,000 5,000 4,000 3,000 2,000 1,000 0
Suppliers (owed) 0 0 0 0 0 0 0
Working capital 6,000 5,000 4,000 3,000 2,000 1,000 0
Change in working capital (6,000) 1,000 1,000 1,000 1,000 1,000 1,000
General rule from the sheet

The sum of the changes in working capital must be zero at the end: all the cash tied up is eventually released. Use this to check your table.


Quick finder