In late 2018, Goldman Sachs downgraded Analog Devices from neutral to sell. Morgan Stanley followed weeks later. The stock had slid 15% from its high. But one fund manager did the opposite — he bought more.
His valuation models said the stock was worth $96. The market said $87. Eighteen months later, ADI hit $120.
How did he see what Wall Street missed?
In Post 6, we learned one valuation method — DCF. It’s the workhorse of finance, but professional analysts never rely on just one number. They triangulate. They run multiple methods and compare. When three different approaches agree, confidence goes up. When they diverge, that’s where the real insight lives.
In this post, you’ll learn two more valuation methods — the Abnormal Earnings Model and Multiples-based valuation — and then use all three to make a real investment decision.
We’ll also meet a new company: Analog Devices (ADI), a semiconductor company so dominant that buying more of its stock when Wall Street was running scared turned out to be the right call.
- You want to understand how analysts actually make investment decisions
- You’ve read Post 6 and want to go beyond DCF
- You want to learn when valuation methods lie — and what to do about it
So far in this series, we’ve used Intel as our case study — a giant trying to reinvent itself through massive factory investments. Now we’re switching to a very different kind of semiconductor company.
If Intel is like a restaurant chain trying to reinvent itself — huge kitchens, massive scale, betting billions on a new menu — then Analog Devices (ADI) is like a boutique restaurant with a Michelin star. Small menu. Premium prices. Customers lined up around the block. And almost no competition.
Analog vs. Digital: Two Different Worlds
To understand ADI, you first need to understand the difference between analog and digital semiconductors. They’re both chips, but they do completely different jobs:
- Process continuous real-world signals (sound, temperature, voltage)
- Used in sensors, amplifiers, data converters, automotive systems, power supply chips
- Competitive advantage comes from design, not manufacturing
- Less cyclical, less capital intensive
- Custom-built for specific applications — high switching costs
- Process binary inputs (0/1 — power on or off)
- Used in processors, memory, GPUs — computational applications
- Competitive advantage comes from manufacturing (smaller transistors = faster)
- Highly cyclical, extremely capital intensive
- More standardized — easier to switch suppliers
The key insight: analog chips require careful calibration to real-world applications and involve a lot of trial and error. Once a customer designs an analog chip into their product, switching to a competitor is extremely expensive and risky. This gives analog chip makers like ADI substantial pricing power.
ADI’s Business Model
Founded in 1965, ADI is a global semiconductor company renowned for its expertise in analog, mixed-signal, and digital signal processing technologies. Its products include data converters, amplifiers, RF ICs, power management solutions, and sensors — serving markets from automotive and industrial to healthcare and aerospace.
ADI produces about half its output at its four fabrication facilities (fabs) and contracts the remainder to outside foundries, mostly TSMC. The company builds strong relationships with customers, collaborating closely to develop custom designs for complex engineering challenges.
The Acquisitions That Changed Everything
ADI didn’t become dominant by organic growth alone. Two major acquisitions transformed the company:
Producer of microwave and millimeter-wave semiconductors for telecommunications, military, aerospace, and automotive. Some analysts thought ADI overpaid — but the rise of 5G and automotive radar vindicated the deal.
A major competitor in analog semiconductors, strong in power management chips. Expanded ADI’s position in electric vehicles and IoT while consolidating its lead in amplifiers and data converters.
“Combining these two companies was like LeBron James and Kevin Durant teaming up. The two companies had the highest margins and the best end markets in the analog sector.” — Neal Nathani, Totem Point Management
After the Linear deal closed, only Texas Instruments remained as a meaningful competitor. The analog semiconductor market had essentially become an oligopoly — just three main players: ADI, TI, and Maxim.
Strengths vs. Concerns
- Market leader in high-performance analog semiconductors
- Differentiated player with “know-how” edge over competitors
- Limited competition in oligopolistic industry with high barriers to entry and switching costs
- Diversified customer base, relatively insulated from business cycles
- Well-positioned for 5G, IoT, and autonomous driving tailwinds
- Industrial and automotive segments slowing (70% of the business)
- Performance dependent on macroeconomic conditions outside ADI’s control
- May have overpaid for acquisitions ($2.5B for Hittite, $14.8B for Linear)
- RNOA declined dramatically from over 85% (2011) to just over 10% (2018)
- Not clear if acquisitions would produce sufficient returns
This is the backdrop: a dominant company in a shrinking competitive field, with powerful tailwinds — but also a stock that Wall Street was souring on. To decide whether ADI was a buy or a sell, we need more valuation tools beyond DCF. Enter the Abnormal Earnings Model.
The analog semiconductor industry focuses on the design, development, and manufacture of analog integrated circuits (ICs) and related components. It is much less cyclical and less capital intensive than its digital counterpart.
Why? Because analog chips derive their competitive advantage from design rather than manufacturing. While digital chips compete on transistor size (who can make the smallest, fastest chip), analog chips compete on how well they can handle real-world signals. This requires careful calibration and lots of trial and error.
The result is a highly concentrated, semi-oligopolistic market with just a few major players. Analog semiconductor purchasers are generally willing to pay a premium for high-quality, long-lasting, custom-built products. Once a customer designs an analog chip into their device, switching costs are extremely high — they’d have to re-engineer the entire product.
This dynamic gives analog chip makers substantial pricing power and helps explain why ADI could maintain high margins for decades — and why analysts believed those margins could sustain or even expand going forward.
Here’s an analogy. Your bank account earns 5% interest — that’s the “normal” return. If you put your money into an investment that earns 12% instead, the extra 7% above the bank rate is the “abnormal” return. You’re earning more than the minimum expected.
The Abnormal Earnings Model (AEM) applies this same logic to an entire company. It asks: how long can this company keep earning above the minimum? And what should investors pay for that ability?
Normal Earnings
At the very least, a company’s earnings should equal the cost of the money needed to fund and run the business. If it can’t even do that, the business will eventually fail. This baseline is called normal earnings:
Abnormal Earnings (Residual Income)
Anything a company earns above normal earnings is called abnormal earnings, also known as residual income. It’s the excess — the earnings that go beyond merely covering the cost of equity:
As Professor Srinivasan explains: “Investors would be willing to pay a premium over book value for the stock of a company only if it is able to earn more than the expected rate of return on book value — in other words, normal earnings. Excess earnings or abnormal earnings is the capacity of the firm to generate above the normal or expected level.”
The AEM Formula
The Abnormal Earnings Model values a company as two pieces:
Think about what this means:
- If a company is expected to earn only normal earnings (abnormal earnings = 0), investors should only be willing to pay book value for the stock.
- If abnormal earnings are positive, investors will pay a premium over book value — the company is creating value beyond the minimum.
- If abnormal earnings are negative, the stock should trade below book value — the company is actually destroying value.
The Big Advantage Over DCF
In Post 6, we saw that DCF puts enormous weight on the terminal value — the estimated value of all cash flows beyond the forecast horizon. For Intel, the terminal value was the single biggest piece of the valuation, and it was also the most uncertain.
AEM fixes this problem. By separating the valuation into book value (known and predictable) and abnormal earnings (uncertain but smaller), the terminal value component has less weight:
+ Terminal Value (HUGE)
+ PV of Abnormal Earnings
+ Terminal Value (smaller)
As the course material puts it: “AEM ameliorates the problem that DCF has with a larger weight on terminal value by separating out the part of the estimate — the book value of equity — which has a more easily predictable rate of return across companies.”
Mean Reversion: Why Abnormal Earnings Don’t Last Forever
Here’s a fundamental economic principle: competition drives out excess profitability. Companies with high abnormal earnings attract competitors who try to copy their business models. Over time, those excess profits get competed away.
The data bears this out. When researchers tracked abnormal earnings across thousands of public companies over 10 years, they found a striking pattern: all companies — whether they started with very high or very low abnormal earnings — converged toward zero by year 10.
But some companies can sustain abnormal earnings longer than others. What makes the difference? Sources of abnormal earnings include:
- Competitive advantage through differentiated products
- Cost or efficiency advantage from advanced technology or economies of scale
- Network effects — the product becomes more valuable as more people use it
- First-mover advantages — the first company to establish a market position
- Superior management quality
- Patents and know-how — legal protection of innovation
ADI has several of these: differentiated products, cost advantages, high switching costs, and an oligopolistic market structure that limits competition. This is exactly why analysts at Totem Point believed ADI could sustain abnormal earnings longer than the average company.
What Should an Analyst Consider?
To forecast how long a company can maintain abnormal earnings, an analyst should consider:
- Barriers to entry — and how long they might last
- Competitive pressure by geographic market — which markets are mature vs. still growing?
- Supplier and supply chain pressure — potential shortages or production backlogs?
- Anticipated organizational changes — mergers, divestitures, and their impacts
- Quality of assets — will goodwill write-downs be required?
- Macroeconomic trends — the broader economic environment
- Recent performance — what drove success (or failure), and will those conditions continue?
The AEM is derived directly from the Discounted Dividend Model (DDM). Here’s how.
Start with a basic accounting identity: the change in a company’s book equity at the end of a period equals the difference between net income and dividends:
Rearranging to solve for dividends:
Now substitute this into the DDM formula. When you do the algebra, the dividend terms transform into book value plus abnormal earnings terms. The intrinsic value becomes:
For a two-period example:
As the forecast horizon expands, the final term — the PV of book equity at the end of the last forecasting period — becomes very small (the denominator becomes very large). What remains is just book value plus the present value of future abnormal earnings.
This is why DCF and AEM give identical results: they’re both derived from the same underlying DDM formula, just rearranged differently.
Here’s the punchline: when we apply both DCF and AEM to Intel using the same forecasts from Post 6, we get identical results.
Why identical? Because both methods are derived from the Discounted Dividend Model and use the same assumptions to forecast earnings and cash flows. The difference isn’t in the answer — it’s in how each method frames the calculation:
- DCF bundles everything into future cash flows. It asks: “what will the company generate?”
- AEM separates book value (known) from abnormal earnings (uncertain). It asks: “how much value are managers adding beyond the minimum?”
Same answer. Different confidence profile. AEM tells you where the value creation actually comes from.
Three Drivers of Abnormal Earnings
Professor Yuan Zou identifies three competitive forces that drive abnormal earnings:
Through differentiation — reflected in profit margin. Example: Target carries higher-priced, better-quality products than Walmart. Target has higher margins because customers pay more for the experience.
Through efficiency — reflected in asset turnover. Example: Walmart is famous for pushing suppliers to lower prices and arranging just-in-time delivery. It’s very efficient at turning assets into sales.
Through intangibles — innovation, patents, strong brands. Companies like Coca-Cola and Google maintain competitive advantage through brand power and intellectual property.
The P/B Ratio — What AEM Tells Us About Price-to-Book
A special case of the AEM is the steady-state model, where revenue, net income, and book value all grow at a constant rate so that ROE remains constant. In this case, the Price-to-Book ratio simplifies to a beautiful formula:
This simple equation tells us two powerful things:
And here’s the critical insight from Professor Zou: “Growth can destroy value if ROE is lower than the required return. If the company cannot deliver return more than the required return by shareholders, growth could hurt value.”
This is counterintuitive. We usually think growth is good. But if a company is growing while earning less than its cost of equity, every dollar of growth actually makes the company less valuable. It’s like a restaurant that loses money on every meal — serving more customers just makes the losses bigger.
DCF is the most widely used method and works for virtually all companies. It’s the go-to tool when you have good cash flow forecasts. However, it puts a heavy weight on terminal value, which is the most uncertain part.
AEM reduces the terminal value weight, making it better when book equity is a meaningful starting point. It’s particularly useful for companies with significant tangible assets (banks, manufacturers, utilities) where book value is a good anchor.
Both methods give the same answer when using the same inputs. The practical difference is what each method highlights: DCF focuses on cash generation, while AEM isolates how much value managers are adding beyond the minimum expected return. Professional analysts often run both as a cross-check.
You want to price your house. You could build a detailed model of future rental income, discount it back to today, and estimate the present value. That’s essentially DCF.
Or you could just look at what similar houses in your neighborhood sold for last month. That’s multiples.
As Professor Yuan Zou explains: “Investors and analysts usually use the multiples approach to value a company. This approach is used because no single method is comprehensive, and analysts use a combination of methods to triangulate and test their valuation.”
The core assumption is simple: similar assets should be sold at similar prices. If comparable companies trade at 20 times earnings, your company probably should too — unless there’s a good reason it shouldn’t.
The Four Multiples
“How much investors pay for each dollar of earnings.” The most popular ratio — data is easy to acquire.
“How much investors pay for each dollar in net assets.” Useful when assets primarily drive earnings.
“How much investors pay for each dollar in sales.” Very useful for firms with negative earnings — young and growth companies.
“Mostly represents the operating decisions of the firm.” Some analysts prefer this to P/E as it strips out financing and tax differences.
Worked Examples
Example 1: Price-to-Book. Imagine Small Fry Company had $8 million in book equity, 100,000 shares outstanding, and opened at $124 per share. What’s its P/B?
Book equity per share = $8,000,000 / 100,000 = $80
P/B = $124 / $80 = 1.55
This suggests investors believe the company will generate future earnings above its book value and cost of equity. They’re paying a 55% premium over the accountant’s estimate of value.
Example 2: EBITDA Multiple. A company has EBITDA of $160 million, share price of $190, and 8 million shares outstanding. Peers trade at a weighted median EBITDA multiple of 10.04x.
EBITDA per share = $160M / 8M = $20
Implied value = $20 × 10.04 = $200.80 per share
Current price: $190 → undervalued by $10.80 (~5.7%)
But before you rush to buy, ask: Are peers truly comparable? Is this company much smaller or larger? How does its capital structure compare? What explains the market’s lower valuation?
The 5-Step Process
- Select a performance measure (earnings, sales, book equity, cash flows) as the basis for the multiple
- Identify comparable benchmark companies — typically in the same industry, similar size and growth
- Calculate peer multiples — divide each peer’s price by the performance measure
- Apply the average peer multiple to your company’s performance
- Compare the calculated value to the actual share price — does it make sense?
Quick Check: P/E Facts
That last point matters. P/E uses current earnings, which can be distorted by one-time events. As we’ll see with ADI, a single tax law change threw off the P/E calculation by enough to mislead analysts who weren’t paying attention.
Comparing house prices works great on the same street. Same neighborhood, similar square footage, comparable lot size — the comps make sense.
But comparing a Manhattan penthouse to a farmhouse in Iowa? Same “number of bedrooms” doesn’t make them comparable. Multiples have the same problem.
As Professor Zou warns: “The pitfall is that it can be too simplistic. It is actually very challenging to come up with good comparables, and the economics of the focal company will be different from other companies, even for carefully selected peers.”
The Five Complications
Even companies in the same industry have different business models. As Srinivasan notes: “Even across firms in the same industry, there are differences in business models that are likely to affect their multiples.” ADI and Intel are both semiconductor companies, but one makes analog chips through design expertise and the other makes digital chips through massive manufacturing scale.
When a company has negative earnings, P/E breaks completely. A negative P/E ratio is meaningless and incomparable. You can’t say a company with P/E of -15 is “cheaper” than one with P/E of -30. Book value multiples (P/B) are “less subject to temporary distortions” and may still work.
Multiples don’t account for capital structure. Two companies with identical operations but different debt levels will have different P/E ratios. The heavily leveraged company may appear “cheaper” on P/E simply because interest expense reduces its earnings, not because its operations are less valuable.
When a company is in flux — acquiring, restructuring, pivoting its business model — its multiples are temporarily distorted. Intel during its IDM 2.0 transformation (from Post 5) is a perfect example: its current earnings didn’t reflect its future potential, making P/E misleading.
Major acquisitions can lead to “temporary and significant alterations in performance and the debt/equity structure.” ADI’s acquisition of Linear Technology for $14.8B dramatically changed its balance sheet and depressed its RNOA from 85% to 10% — not because operations worsened, but because the asset base ballooned.
The P/B Connection to Fundamentals
Despite these complications, multiples aren’t disconnected from fundamental analysis. Srinivasan identifies the three main drivers of the price-to-book multiple:
- Profitability (ROE) — a superior ROE and the ability to generate excess profits over cost of equity
- Growth rate (g) — growth in revenues and the capacity to reinvest in the business
- Ability to sustain both — competitive advantages that protect profitability and growth over time
The critical rule: “Growth only adds value if the return on equity is higher than the cost of equity.”
When Growth Breaks the Formula
What happens if the growth rate (g) exceeds the cost of equity (re)? The steady-state formula P/B = 1 + (ROE − re) / (re − g) produces a negative denominator, giving a nonsensical result.
As the course material explains: “If the growth rate of equity was greater than cost of capital, the company would eventually take over the world!” No company can sustain a growth rate above its cost of capital indefinitely. Small companies can maintain high growth for extended periods, but they’ll eventually attract competitors. Long-term, book equity growth rates converge to the overall economy’s growth rate.
This is why practitioners, as Srinivasan emphasizes, “consider several methods simultaneously to validate their valuation results. They would follow a multiples-based approach supplemented by a valuation model, either a DCF or an abnormal earnings model.”
Starting from the steady-state AEM formula where ROE remains constant and book equity grows at rate g:
Divide both sides by B0 to get the price-to-book ratio:
The inverse (book-to-price ratio) is also useful:
Notice the linear relationship between P/B and ROE: as ROE increases, P/B increases proportionally. This is the formal connection between a company’s operational performance and its market valuation multiple.
Key constraints: the formula only works when re > g (otherwise the denominator is negative). And it assumes steady-state conditions — in reality, ROE and growth rates change over time.
It’s late 2018. You’re the portfolio manager at Totem Point Management. Here’s the situation:
- Bought ADI at $55/share in June 2016
- Stock now at $87/share — nearly 60% return in 2.5 years
- Totem’s target: 50% return over 2-year horizon — already hit
- Goldman Sachs downgraded ADI: neutral → sell
- Morgan Stanley downgraded: overweight → equal weight
- Stock down 15% from June 2018 high
Totem Point’s Analysis
Rather than following the crowd, Totem Point ran a rigorous analysis. Their valuation assumptions:
| Assumption | Value |
|---|---|
| Beta | 1.25 |
| Risk premium | 5.5% |
| Risk-free rate | 3.10% |
| Cost of equity | 10.0% |
| Historical P/E multiple | 18 |
Their key forecast assumptions:
- NOPAT Margins: Low-to-mid 20% range, with some expansion expected (increasing pricing power)
- Capital Structure: Stable — no planned changes to net debt / net operating assets ratio
- Asset Utilization: Largely unchanged — efficiency gains would be offset by customization costs
Revenue Forecasts by Segment
Totem built a detailed 10-year revenue forecast by analyzing each of ADI’s four business segments individually:
| Segment | 2019 | 2021 | 2023 | 2025 | 2028 |
|---|---|---|---|---|---|
| Industrial | 16.0% | 10.0% | 8.0% | 6.0% | 3.5% |
| Automotive | 3.0% | 5.0% | 10.0% | 10.0% | 3.5% |
| Communications | 19.0% | 45.0% | 55.0% | 35.0% | 5.0% |
| Consumer | -3.0% | 4.0% | 4.0% | 3.5% | 3.5% |
| Implied Total | 11.9% | 17.3% | 25.6% | 22.0% | 4.5% |
The story was clear: Totem believed 5G deployment would drive explosive growth in ADI’s Communications segment, more than compensating for the slowing Industrial and Consumer segments. Autonomous driving would provide a modest boost to Automotive.
The Three Valuations — All Converge
Using these forecasts, Totem ran all three valuation methods. The result was remarkable:
Three independent methods, one answer: ADI was worth roughly $96–$97. The market was pricing it at $87. Someone was wrong.
As the course notes: “Clearly, Totem and the market are in disagreement about the future of the company and its industry. The implied market assumption is either that cost of capital will increase or that ADI will face greater competition going forward, or both.”
Here’s the analyst’s framework for interpreting this: “In general, larger abnormal profitability is consistent with a more significant strategic moat and lower competitive threat, and smaller abnormal profitability is consistent with greater competition. The ROEs and RNOAs implied by a forecast should be consistent with the analyst’s views about how industry competition is expected to unfold and how the company’s competitive edge is expected to change.” Totem believed ADI’s oligopolistic market and 5G tailwinds would sustain its abnormal profitability — the market disagreed.
Your Turn: What Would You Do?
Srinivasan’s Takeaways from the ADI Case
- Valuation is always accompanied by uncertainty — particularly for growth companies that can grow at significantly faster rates than the overall economy
- Deep content and industry knowledge is critical — analysts conduct in-depth field research to develop clarity about how economic and technological trends will influence growth prospects
- Don’t get emotionally attached to a thesis — investment managers develop rigorous frameworks, like Totem Point’s “idea matrix,” to ensure decisions are informed by best available facts, not emotions
- Understand the range of plausible intrinsic values — reconcile why valuation estimates differ across different approaches. An analyst’s contextual understanding helps inform subjective probabilities of upside and downside scenarios
- Be mindful where you disagree with market consensus — and why — this helps check your own information sources and biases
- Corporate managers should understand how investors work — ADI managers can tailor their disclosure practices to better serve investor needs
Totem used a historical P/E of 18 as a cross-check against their DCF and AEM valuations. But they had to adjust for a one-time distortion.
The Tax Cuts and Jobs Act of 2017 (effective January 1, 2018) forced ADI to pay a one-time tax of $70 million on repatriated earnings of its foreign subsidiaries. This temporarily reduced 2018 earnings.
P/E multiples are particularly sensitive to transitory shocks in earnings. A one-time $70M hit significantly influences current period net income but has a relatively minor effect on stock prices because investors know the shock is temporary.
By adjusting ADI’s net income — adding back expenses not closely related to core operations — Totem produced an earnings number more representative of operating performance. The adjusted multiples valuation came out to $96.63, validating the $96.66–$96.71 range from DCF and AEM.
This is exactly why analysts triangulate: each method catches what the others might miss.
We’ve now covered all the valuation methods a professional analyst uses. As Srinivasan summarizes: “While there are different approaches to conducting a valuation exercise, they all derive from the same underlying model. Practitioners often consider several methods simultaneously to validate their valuation results.”
Here’s the complete picture:
| Method | What It Values | Benefits | Caveats |
|---|---|---|---|
| Discounted Dividends (DDM) | PV of future dividends | Reflects the basic understanding that a company’s value today is based on its future returns to shareholders. | Many companies don’t pay dividends regularly. Returns come as share price appreciation instead. |
| Discounted Cash Flow (DCF) | PV of free cash flows | Reconfigures returns to shareholders as future cash flows to equity. Most widely used method. | Puts a heavy weight on terminal value — the most uncertain part of the calculation. |
| Abnormal Earnings (AEM) | Book value + PV of abnormal earnings | Reduces terminal value weight (book value captures a portion). Allows for a more precise understanding of abnormal profits and their source. | Estimates of the expected return on book value must be accurate. |
| Multiples | Peer comparison ratios | Simplifies calculations. No need to project future earnings. Great for triangulation and sanity checks. | Relies on market pricing peers correctly. Peers must be truly comparable. |
Key Connections
Srinivasan draws important connections between the methods:
- The P/B multiple is a function of future abnormal ROEs, book equity growth, and the firm’s cost of capital
- The P/E multiple is a function of those same three factors plus current return on equity
- Multiples are popular because they don’t require multiyear forecasts of performance
- But “even across firms in the same industry, there are differences in business models that are likely to affect their multiples”
The practical takeaway: “They would follow a multiples-based approach supplemented by a valuation model, either a DCF or an abnormal earnings model.”
Laura Champine on Analyst Biases
Laura Champine, a veteran sell-side analyst, identifies several biases that affect valuation in practice:
“We are infamous for building in for a growth company... an acceleration of the growth rate, and that might not have a macroeconomic overlay.” Sell-side analysts tend to project growing companies will grow even faster, without considering whether the broader economy supports that assumption.
“This is a very famous issue. There have been settlements around this with the major investment banks.” Analysts need good relationships with management teams to get information. This creates a natural positive bias — you don’t want to write negative reports about people you need access to.
For mature or value stocks, analysts tend to build in “just kind of a natural tail off in growth” without analyzing company-specific drivers. Champine’s antidote: “It’s very important to think of each company as an individual entity and look at its company-specific drivers rather than just making another model.”
Champine’s core philosophy: “It’s not a soothsayer’s job. I’m not playing craps in Vegas. An Excel spreadsheet is only as good as its inputs. How can I use this mosaic of information that I’ve tried to take in for the last two months and the last 20 years and make some sense as to the likely growth rate, with so many different overlays — macro overlays, secular overlays, company-specific overlays?”
Champine reveals an often-overlooked dimension of analyst work: understanding management psychology matters as much as the numbers.
“Some management teams are always conservative, and so the thought process is that they’ll always beat expectations. They might have a really good quarter, and the stock might trade down on that.”
Why would a stock fall after a great quarter? Because if management always under-promises, the market expects them to beat. So merely meeting high expectations isn’t enough — the stock only moves on surprises relative to what the market already baked in.
Champine’s advice: “I have to understand that part of it, too — what do people think about this management team, not just from a business perspective but from a psychological perspective?”
And her parting wisdom: “There’s so many layers to the business, and it changes every day. One thing I can say for what I do is that it’s never boring, and I’m always talking to smart people.”
Looking Ahead
With all four valuation methods in our toolkit, Srinivasan bridges to the next challenge: “How should we value multi-segment businesses? Companies often grow by buying other businesses. How should we value acquisition targets?”
Consider Disney: it competes with Netflix in entertainment, but it also runs theme parks. How do you value a company that operates across completely different industries?
When a company like Disney operates in both entertainment (competing with Netflix) and theme parks (a completely different business), a single set of multiples won’t work. Entertainment companies trade at very different multiples than hospitality companies.
The solution is sum-of-the-parts valuation: value each business segment separately using peers and multiples specific to that segment, then add the values together. This is particularly important for conglomerates and companies considering spinoffs.
In future posts, we’ll explore this through real M&A cases: how companies value acquisition targets, when to spin off a division, and how the market reacts to these decisions.
A company has $200M book equity, 8% cost of equity, and earns $22M net income. What are its abnormal earnings? And if abnormal earnings persist at this level forever, what would investors be willing to pay for the company?
Normal earnings = 8% × $200M = $16M
Abnormal earnings = $22M − $16M = $6M
If perpetual: Value = Book equity + Abnormal earnings / cost of equity = $200M + $6M / 0.08 = $275M
Investors would pay a $75M premium over book value for the company’s ability to generate excess profits. In reality, abnormal earnings mean-revert toward zero, so the actual premium would be lower than $75M.
Two companies in the same industry: Company A has a P/E of 25, Company B has a P/E of 12. An intern concludes that Company B is undervalued. What three questions should you ask before agreeing?
Before concluding B is undervalued, ask:
- Are they truly comparable? — Even in the same industry, different business models, sizes, or geographies can justify different multiples. A luxury retailer and a discount retailer are both “retail” but deserve very different P/Es.
- Does Company B have higher leverage? — More debt means more interest expense, which reduces earnings and inflates the P/E comparison. B’s lower P/E might just reflect higher debt, not better value.
- Is Company B in a transition period? — If B recently made a large acquisition, restructured, or had a one-time earnings boost, its current P/E may be temporarily distorted and not representative of ongoing performance.
Your DCF says a stock is worth $50. Your AEM says $50. But multiples say $35. What’s the most likely explanation?
Two likely explanations:
- Peers may be undervalued by the market. Multiples rely on peer prices being “correct.” If the entire sector is depressed (sector-wide pessimism), the multiples will suggest a lower value even for a strong company.
- Your company may genuinely be better than its peers in ways multiples can’t capture. DCF and AEM use company-specific forecasts and assumptions. Multiples use market consensus. If your company has a superior moat, better management, or stronger growth prospects, its fundamental value (DCF/AEM) should exceed the peer-average multiple.
The $50 DCF/AEM agreement gives you confidence in the company-specific analysis. The $35 multiples tell you the market doesn’t see it yet — which could be an opportunity (like ADI at $87) or a warning that you’re missing something.
It’s 2024. ADI stock is at $220. The original Totem Point analysis valued it at $96 in 2018. Using ROIC and mean reversion logic, what should you check before deciding whether to buy at $220?
Before buying at $220, check:
- Has RNOA/ROIC reverted toward cost of capital? Mean reversion predicts that abnormal earnings converge to zero over ~10 years. If ADI’s ROIC has declined toward its WACC, the premium over book value should be shrinking.
- Are abnormal earnings declining? Check if the spread between ROE and cost of equity has narrowed since 2018. If it has, the stock’s premium over book value is harder to justify at $220.
- Is the P/E still reasonable vs. the historical 18x? If the P/E has expanded well beyond 18, the market may be pricing in unrealistic growth expectations.
- At $220, what growth rate is implied? Work backwards from the stock price to calculate what growth rate the market expects. If it implies growth significantly above GDP or above ADI’s cost of equity, the formula (and the stock) may be unsustainable.