[2025–2026] Week 2: Benjamin Graham’s Security Analysis, Part 2
Reading assignment and questions for week 2 of the Value Investing Seminar
I was happy to see how many of you engaged with the reading and the questions in a thorough, high-quality way. Keep it up – while this is not easy reading, it will form the foundation of your investment process. To give you a little encouragement: if you make it past this week’s reading assignment, it gets easier from there.
I just wrote an article answering question 0 from last week’s assignment about the dimensions of an investing style where I list 21 important dimensions. Please take a look at the article, and if you find it helpful, please consider restacking it to help grow our community.
Switching over to Security Analysis, Part 1 and Benjamin Graham, many of you correctly noted that the first edition was written during the Great Depression and not long after the stock market crash of 1929. That had a big impact on both Graham (whose partnership declined a meaningful amount in part due to use of margin debt) and on overall investor psychology.
Graham was not writing to today’s audience of investors who have been conditioned that stocks only go up over the long-term. He was writing to people who had just seen an unprecedented collapse in stock prices and many of whom think that purchasing any stock is a speculation. That is why he felt the need to work so hard to convince his readers that stocks can be an investment and why he laid out such a careful, risk-averse approach to investing.
Before we move on to Part 2 and this week’s assignment, here are some excerpts from some of your answers to last week’s questions that I thought are worth sharing with the community. There were far more insightful answers than I can fit in here, and over time I will try to highlight as many folks’ contributions as possible:
In response to Question 1 Trishit wrote: “The environment around 1933 was characterized by a double discrediting of serious analysis: 1. Persistence of imaginary values [prior to the crash] (pp [61]) and 2. Disappearance of real values [after the crash] leading to devasting effects on established earning power.”
In response to Question 2 Atreya wrote: “Graham’s investment philosophy focuses on finding intrinsic value using definite standards, and purchasing securities at a bargain to the intrinsic value. He thinks its best by comparing it to the chances of failure of other approaches such as market timing- which cannot be done over and over again, or the “new era theory” which became prevalent after 1927, where a study of fundamental aspects of securities were abandoned in favour of spurious metrics espoused by stock promoters. Graham also notes that an investment is for permanent holding. Therefore, most people who buy securities, in the hope of selling it at a later date at a higher price are speculators. Graham’s assertion is that this is the only sensible way to invest as this can assure safety of principal under most conditions and has a reasonable probability of delivering an adequate return.”
In response to Question 3 George wrote: “I admire Graham’s clarity of thought, he knows and can clearly explain what the necessary conditions are for something to qualify as an investment. It is also strongly implied that he has a robust process, perhaps a checklist for evaluating investments. I would benefit from defining my investment philosophy more clearly and bringing more structure to my process.
One area where I suspect that an issue might arise is in finding investments in businesses with ‘inherent stability’ that are priced cheaply in absolute terms. I don’t disagree that these investments tend to deliver good returns, but much of the time such businesses are priced to deliver (see Costco, Autozone currently, for example). There are avenues available to mitigate- compromise on the degree of inherent stability required, accept a lower margin of safety or be able to endure long periods of inactivity/allow uninvested cash to pile up.”
In response to Question 4 Alberto wrote: “Graham defines investment as an operation that, after careful analysis, promises safety of principal and an adequate return. Anything that doesn’t meet these criteria is considered speculation.
The implication, as I see it, is that a careful analysis must be conducted first—one grounded in facts that are knowable and relevant—to determine whether the investment will preserve the principal and offer a return that is adequate compared to prevailing alternatives.”
Regarding last week’s Question 4, I am going to add a thought question: What did Graham mean by “an investment operation” in his definition above? As a corollary, can a purchase of a single security in and of itself be classified as either an investment or a speculation according to Graham? According to you?
Week 2 assignment is to read Part 2 of Security Analysis and answer the following questions:
Question 1: Why does Graham believe that investing in High Grade fixed income securities to be a ‘negative art’, in contrast to investing in common stocks? What does he mean by that? Do you agree?
Question 2: What is Graham’s main idea about what provides safety in a high-grade bond? Do you agree?
Question 3: How does Graham evaluate the safety of a high-grade bond based on the company’s financials? Is there anything you would do differently with what you know today? Why?
Question 4: Which security structures does he believe to be permanently inferior in this category? Why?
Question 5: Miniature Credit Analysis scenario:
- Cab driver has driven a cab for someone else, and now wants to buy his own for $100
- The cab will have a useful life of 5 years, and have straight-line depreciation with no residual value for accounting purposes
- For cash purposes, the cab will have no cash outlays associated with it but will need to be replaced at the end of 5 years, with another cab, which we will assume will also cost $100
- The cab driver has historically averaged EBIT of $10 on his historical revenues of $100, and expects that to continue. There are no taxes.
a) What questions would you like to ask/what would you like to know to assess whether buying a 5-year $100 bond with a 5% coupon meets Graham’s standard for investment in a high-grade bond?
b) Substantiate your answer to a) with specific credit metrics that support your analysis. What metrics are most relevant? Why?
c) What would change in your analysis, if anything, if everything staid the same but the bond had a 3-year maturity? A 7 year maturity?
d) How much would it matter to Graham whether the bond was unsecured or if it were secured by the cab? What about you?
Question 6: Find and analyze two bonds using both Graham’s process and any of your own modifications that you think are reasonable. Provide your analysis of:
a. A bond rated BBB or BBB- by S&P with a maturity at least 5 years away
b. A bond rated A or higher by S&P with a maturity at least 5 years away
Is either of those two bonds a good investment in a) Graham’s view b) your opinion? Why?
Question 7: See if you can think of an AI prompt based on Part 2 of Security analysis that would make your work easier. Suggestion: use a thinking model (e.g. Gemini 2.5 Pro or ChatGPT 5; if the latter set reasoning_effort=high; consider whether you need “Deep Research” mode enabled or not for what you are trying to do).
Now it’s your turn:
Submit your answers in the comments below this article with all your answers in a single comment. I will engage with some of the answers each week and highlight some of the ones I find most insightful in next week’s seminar assignment article.
Engage with the answers of some of your fellow seminar members in the comments below. Remember – the goal is to learn together. Be kind, be respectful and try to add to our learning as a community.
Feel free to ask any questions about the reading in your comment.
Until next week,
Gary
About the author
Gary Mishuris, CFA is the Managing Partner and Chief Investment Officer of Silver Ring Value Partners, an investment firm that seeks to apply its intrinsic value approach to safely compound capital over the long-term. He also teaches the Value Investing Seminar at the F.W. Olin Graduate School of Business.
1. In fixed income investing, upside is capped. Downside is not. This can result in loss of principal. Therefore, to reduce loss of principal at a portfolio level, Graham says that one should avoid losers. I fully agree with him. In fixed income investing, errors of commission (Type 1 error) must be avoided, while errors of omission (Type 2 error) don’t represent large opportunity costs. This contrasts with venture capital investing where Type 2 error significantly hurts portfolio level returns compared to Type 1 errors.
2. Graham laid out a few general principles to ascertain safety of bonds.
a. Ability to repay is far more important than security over assets
b. Ability to repay should be considered in depression conditions
c. We should not sacrifice safety for higher yield
d. Larger companies are safer
e. More stable industries are safer
f. Track record of payment of interest and dividends over a long period of time
g. Track record of management performance over a long period of time in the past
h. Safety should be tested against well-defined standards
3. Graham recommends certain specific tests to determine the safety of high grade bonds. I’d modify these standards by making the tests even more severe. There is little penalty for this. There are many bonds. Yields are not that different. My pessimism is rooted in the fact that we have not lived through a prolonged depression condition over the last 10-15 years- hence a “laboratory-test” of earnings power under adverse conditions is unknown.
a. Earnings Coverage Over Fixed Charges: Graham’s recommends a ratio of 3x for industrials, and 1.75x for utilities over the last 7-10 years. I’d look for securities where ((EBITDA-maintenance capex-lease payments)/interest) ratio of at least 5x in each of the last 10 years for market-oriented businesses, and atleast 2x in each of the last 10 years for regulated businesses with contractual cash flow streams (gas pipelines, electricity transmission, etc). I’d also look for businesses whose (EBITDA-maintenance capex-lease payments) is atleast 80% of its previous high (in last 15 years)
b. Equity/Debt ratio: Market value of equity/total market value of debt should be atleast 1 for equities and 0.67 for utilities. I’d make this more stringent through two ways: Using the ratio (tangible book value of equity/market value of debt) if tangible book value of equity < market value of equity, and using the ratio (market value of equity/market value of debt) if tangible value of equity > market value of equity. And making this 1.5x for all kinds of securities.
c. Working Capital ratios: Current assets/current liabilities > 2. I’d modify this by making an exception for exceptional consumer businesses like Haier Smart Homes, and Nestle India which meet criteria 1 & 2, but have negative working capital since their businesses are exceptionally dominant. They can demand better terms from their customers and suppliers.
4. Graham believes that the typical preferred stocks to be permanently inferior to high grade bonds.
This is because high grade bonds have promised repayment and priority of repayment over preferred stocks and equity. In contrast, preferred stock have no promised repayment, and only has priority over equity.
However, Graham concedes that exceptional preferred stocks have similar characteristics as high grade bonds (eg: Nabisco preferred stock in the 1930s).
5a:
• Highest and lowest EBIT in the last 10 years. Compare it to the $5 interest rate.
• Sensitivity of EBIT to fuel prices
• Effects of ride hailing on EBIT of cab in industry
• What is the recourse for the bond holder if the cab is non-operational due to accident- is there insurance, how does the insurance protect the bond holder?
5b:
• Interest coverage ratio (EBIT/interest)
• Accumulated Cash + residual asset value always > principal
• Tangible book value of equity/debt amount
5c:
If maturity is 3 years, accumulated cash in the enterprise would be $71 (accumulated OCF-maintenance capex (0 in this case)- net change in WC (0 in this case). Therefore, I need to know if the cab can be liquidated after 3 years, at atleast $40+ ($10 margin of safety). If it’s a 7 year bond, then I’d worry about the liquidation value of the cab at 3 years. Accumulated cash by then would be 65 (since 100 will be spent on the new cab). I’d have to ascertain that the cab can be sold for atleast $45 ($10 margin of safety).
5d:
While his primary focus would be on the repayment ability from operating earnings and accumulated cash, he’d like to have security over the bond. I’d too.
6 a: S&P BBB or BBB- --> OXY Corp 7.5 Nov 01’96 with a yield to maturity of 6.5%.
Occidental Petroleum is a large shale producer in the Permian Basin and other oil interests. It produces ~1.4 million barrels of oil and gas/day. It has a market cap of equity of ~$40 billion, and has a net debt of $15 million.
Its debt has recently reduced by $6.5 billion since Buffett bought their subdivision OxyChem for $9.7 billion. Outstanding debt by end of year is expected to be $15 billion. Annual interest expense is likely to be $1 billion.
• At $60 oil prices, depreciation is ~$15/barrel and EBIT ~$10/barrel. If oil price goes down to $40/barrel, exploration expenses can be minimized, and the company can still produce at free cash flows of ~$5/barrel. Company produces ~450 million barrels/year. Therefore even under depression conditions, OXY has a coverage ratio over interest of 2x and in normalized conditions, the coverage ratio is 5x.
• Both tangible book value of equity/total debt and market value of equity/total debt exceeds 2x
• Current assets (including their listed midstream asset)>1.5x current liabilities
• Berkshire owns 30% of Oxy’s equity and can be a backstop in case its not able to meet its obligations (rich daddy).
6b: S&P A or above -->BNSF Corp 7.25 Aug 01’97 with a yield to maturity of 5.6%
BNSF is one of the largest North American railroads, owned by Berkshire Hathaway.
• In each of the last 10 years, (EBITDA-Maintenance capex-net change in working capital) has exceeded 6.
• Tangible value of equity/market value of debt far exceeds, and market value of equity/market value of debt far exceeds 2.
• It is a division of Berkshire Hathaway- an enterprise with an outstanding record of compounding book value, as well as management reputation of honouring its obligations. Its current assets far exceed its current liabilities (if we exclude float- which will not be called at once).
• Maturities are far lower than cash and marketable securities on the balance sheet
Both would be decent investments according to Graham, however not according to me. Bond investments are inherently investments in currencies. For a long term bond to be a satisfactory investment which would retain purchasing power, one needs to be sure that currency won't inflate too much. That requires good behaviour from governments. However, political expediency, across time immemorial, has caused ruling authorities to debase currency (even in ancient rome) at the expense of the savers. US's debt is much higher now compared to Graham's days. Moreover, the current political establishment has thus far been arbitrary with their policies. Thus, expecting these long term bonds to retain purchasing power till maturity may be foolish.
7. I couldn't think of an appropriate AI prompt.
Question 1
Graham regards the choice of bonds as a negative art because the bond is a security with a limited return. In exchange for this limited return the emphasis must be on the avoidance of loss. It is a process of exclusion and rejection of as much risk as possible.
Question 2
The most important thing is the ability to pay, and I agree with this. Trying to collect debt when the debtor can’t pay is always a slow, legalistic and expensive undertaking, no matter how underwritten by guarantees the debt is.
Question 3
He has a selection of check boxes that must be met to qualify for high-grade lending to companies
1) Size – bigger is better. 0.5bn in sales (rough adjustment for inflation)
2) Earnings coverage 4x for industrials, less for more secure infrastructure
3) Poorest year of earnings must still be covered 3x
4) Leverage must be lower than 5x
5) Working capital must be greater than debts
6) Equity value must be at least 75% of debt.
I would still like to see all of these tests met for a secure high grade bond, though some of the tests are very severe by modern standards. Number 3 should be relaxed under some circumstances.
Question 4
Preferred stock, because the security does not stop poor performance in hard times, and limits the upside in good. Either way it’s a poor proposition, of the “heads I win, tails you lose” sort.
Convertible bonds at the discretion of the management of the company. This puts a further limit on upside as if interest rates fall or the market rate of return required falls, they can be redeemed and refinanced rather than rising to higher values.
Question 5
Setting the scenario against Graham’s tests:
1) Very small, so fails as there is clearly a high risk here if the cab driver falls ill, dies or is otherwise incapable of working.
2) Earnings coverage = 10/5 = 2. Not enough.
3) We only know the average so we need to ask what the worst year in the last 7-10 years are according to Graham
4) This test fails as the leverage is 100/10 = 10x. On an EBITDA basis it’s 100/30 = 3.3, but Graham does not use that metric.
5) Working capital we don’t know, so we need to ask. Should be $100, but this is very stringent as this would be all cash as the business needs no working capital according to this model.
6) Irrelevant as it’s not a listed company.
The most relevant metrics are earnings, as these support the whole enterprise. Closely followed by the consistency of earnings as the earnings coverage is not very high. If the cab driver earns more than 5% less than usual he will not be able to fund the interest out of current earnings. Finally the cushion of cash kept in the business is critical in establishing a margin of safety.
c)
If the bond maturity does not match the liability it will require a higher margin of cash in the business, and is therefore less safe. To illustrate this let’s look at the cabbies bank account, assuming he takes no money out of the business. Here are the 3 cases in a table of his bank balance. In the 5 year case his bank balance never falls below $25. In the other two cases it falls to -$25. Hence he will need to keep more working capital in the business to keep it afloat (the margin of safety is reduced)
Business Bank balance
Loan Term 3 year 5 year 7 year
Start 0 0 0
End of year 1 25 25 25
End of year 2 50 50 50
End of year 3 -25 75 75
End of year 4 100 100
End of year 5 25 25
End of year 6 50
End of year 7 -25
d) it would not matter to Graham in his analysis of large corporations, but it would to me in this case. Repossessing a car to repay a debt secured on it is easier, cheaper and much quicker than liquidating a large business, so it makes much more sense, and is therefore more important to secure the debt in this case, especially given the very high risk of this loan.
Question 6
Microsoft Class AAA financials from 06/2025. Values in $bn.
Microsoft 5.2% bond, maturing June 1, 2039
1) Size – 600 times larger than needed in sales
2) Earnings coverage 123/2.3 = 50 times
3) Poorest year of earnings 34/4/2.3 = 14.7 in 2017. NB, growth has rendered this a poor test.
4) Leverage 60/120 = 0.5
5) Working capital must be greater than debts 191 > 60
6) Equity value 3821, debt 60.
An easy pass. But neither Graham nor I would buy it because the yield at 4.88 is only 50 basis points above the nearest UST (May 15, 2039) 4.38 yield to maturity. He does not like taking risk for tiny yield advantage, and neither do I.
Hyatt Hotels Corp.
5.375%, 2031-12-15 S&P: BBB- (stable)
1) Size –6.6/.5 = 13.2
2) Earnings coverage interest/normalised pretax profit 0.411/.18 = 2.3
3) Poorest year of earnings must still be covered 3x. 2020 and 2021 losses due to covid.
4) Leverage 4.11/0.411 = 9.9
5) Working capital must be greater than debts 2.73 < 4.11
6) Equity value must be at least 75% of debt. 13.9 > 4.11*0.75
This one fails on 2, 3, 4 and 5. The current price is above par so the yield is 5.25%. Graham would not buy this one, and neither would I. Hotels are like airlines: when they have full occupancy, they make great profits, but their high fixed costs and semi-discretionary nature make them suffer greatly in recessions.
Question 7
Here is my script for Hyatt Hotels corp. Gives the same answers as me (rather to my relief), but with much more detail. Huge credit to Compound with AI posted by Gary for the basis of the script. I used the ChatGPT 5 Thinking model:
ROLE
Senior forensic bond analyst (ex–Big Four forensic accountant) in professional asset management.
OBJECTIVE
Surface and rank empirically verifiable credit vulnerabilities that could drive loss of principle in bond holding Hyatt Hotels Corp.
5.375%, 2031-12-15
Skip deep, theoretical teardown of the business model or product strategy; focus strictly on red-flag evidence.
DATA-GATHERING
| Bucket | Must-pull docs (last 10 yrs) | Quick query hints |
|--------|-----------------------------|-------------------|
| Core Filings | URD / Doc. de Réf. (AMF), statutory accounts, BALO notices, Banque de France liens (or local equivalents depending on listing jurisdiction) | "Hyatt Hotels Corp" "annual report" OR "rapport de gestion" |
| Transcripts | Earnings calls, broker conferences, investor days | "Hyatt Hotels Corp" "Q&A" |
| Alt-Data | Glassdoor & LinkedIn attrition, trademark/IP disputes, customs/trade data, SimilarWeb traffic, app-store ratings, insider trades, short-interest registers | "Hyatt Hotels Corp" "turnover" OR "employee departures" |
| Opinions | Broker downgrades, ESG notes, activist posts | "Hyatt Hotels Corp" "sell" OR "short thesis" |
Save full URL / file path + page or line # for every excerpt.
---
ANALYSIS Tests (score each finding ratio, Pass, fail)
1) Size – bigger is better. 0.5bn in sales minimum
2) normalised pretax profits/interest costs coverage 4x minimum
3) Poorest year of normalised pretax profits in previous 10 years must still be covered 3x
4) total borrowing/normalised pretax profits must be lower than 5x
5) Working capital must be greater than total borrowing
6) Equity value must be at least 75% of total borrowing.
(No extended business-model section—only include if a fact-based weakness surfaces in the above buckets.)
---
REPORT FRAME (flex length)
1. Executive Summary ≤ 1 page
Give pass/fail conclusion with reasons
2. Ranked Analysis Table (1-6)
| Rank | Test | ratio | difference from minimum | certainty | Pass fail|
3. Detailed Findings
One sub-section per bucket; numbered footnotes link to sources.
4. Falsifiability Checks
What could disprove each core point.
5. Missing-Data Flags
Critical gaps + follow-up query strings.
---
STYLE RULES
• Cite exact file + page for every claim.
• No adjectives without numbers.
• Bullets > prose; zero filler.
• Tables first, narrative second.
---
DELIVERY
Single .docx or PDF dossier. No CSVs, slides, or appendices unless requested.