GrowingGrass.
Back to articles

The Graham Number: What Is It Worth?

Value InvestingNSEInvesting Basics

The previous article introduced two numbers that measure what you actually own when you buy a share: earnings per share and book value per share. These figures tell you what the company is earning and what it has accumulated. But they do not, by themselves, tell you what to pay. A stock with high earnings and substantial book value might still be a poor investment if the price is too high. Conversely, a stock with modest figures might be a bargain if the price is low enough. The question every investor must answer is simple; what is this share worth? To answer it, you need to understand how price relates to the fundamentals.

Price Relative to Earnings and Assets

Knowing a company's EPS and BVPS is useful, but the figures mean little until you compare them to the price you are being asked to pay. Two ratios make this comparison explicit. The price-to-earnings ratio – usually called the P/E ratio – divides the current share price by earnings per share:

P/E Ratio = Price / EPS

If a stock trades at Ksh. 100 and its EPS is Ksh. 10, the P/E ratio is 10. This figure means you are paying ten shillings for every shilling of annual earnings. A higher P/E means you are paying more for each shilling of profit; a lower P/E means you are paying less. All else being equal, a lower P/E is better – you are getting more earnings for your money.

What about BVPS? How does it come to play? The price-to-book ratio – the P/B ratio – divides the current share price by book value per share:

P/B Ratio = Price / BVPS

If a stock trades at Ksh. 100 and its BVPS is Ksh. 50, the P/B ratio is 2. You are paying two shillings for every shilling of net assets. A P/B below 1 means the stock trades for less than the accounting value of its assets – you are buying shillings for less than a shilling. A P/B above 1 means you are paying a premium over book value. Benjamin Graham, the father of value investing, set conservative limits for both ratios. He argued that a defensive investor should pay no more than 15 times earnings (P/E ≤ 15) and no more than 1.5 times book value (P/B ≤ 1.5). Stocks trading above these thresholds carry prices that assume continued growth or market enthusiasm – assumptions that may prove wrong. Stocks trading below these thresholds offer a cushion against disappointment.

Graham went further, combining these two limits into a single formula that provides a ceiling price for any stock. This is the Graham Number.

The Formula

The Graham Number combines earnings per share and book value per share into a single figure:

Graham Number = √(22.5 × EPS × BVPS)

The formula multiplies earnings per share by book value per share, multiplies that product by 22.5, and takes the square root. The result is a price ceiling – the highest price at which the stock can be considered reasonably valued by conservative standards. The number 22.5 comes directly from Graham's two limits; a maximum P/E of 15 and a maximum P/B of 1.5. Multiply them together and you get 22.5. The formula asks: at what price would this stock hit both valuation ceilings simultaneously? That price is the Graham Number.

Applying the Formula

The formula is simple enough. The question is what it reveals when applied to real stocks. Return to KCB Group, whose figures we examined in the previous article. The bank reported EPS of Ksh. 18.70 and book value per share of approximately Ksh. 86 for the financial year ending December 2024(KCB Group, 2025). Plug this into the formula:

Graham Number = √(22.5 × 18.70 × 86) Graham Number = √(36193.50) Graham Number = Ksh. 190

The Graham Number for KCB is approximately Ksh. 190. This price is the ceiling – the maximum price a conservative investor should pay. At the time of writing, KCB trades at around Ksh. 67, which is roughly 35% of its Graham Number, which implies a substantial margin of safety.

The same calculation for Equity Group, with its EPS of Ksh. 12.34 and book value per share of approximately Ksh. 62, yields the following(Group Holdings, 2024).

Graham Number = √(22.5 × 12.34 × 62) Graham Number = √(17212.20) Graham Number = Ksh. 131

Equity's Graham Number is approximately Ksh. 131. With the stock trading around Ksh. 69, it sits at roughly 53% of its Graham Number – also a meaningful discount, though less dramatic than KCB's.

What the Number Tells You

The Graham Number is not a prediction. It does not tell you where the stock price will go tomorrow or next year. It tells you whether the current price is reasonable relative to what the company earns and owns. Unlike many other commodities we procure on a daily basis like a pair of shoes, the value of a stock is not immediately apparent to the buyer. The Graham Number approximates this value into a figure that tells you what a single unit of that business is worth or below which it should trade for. When a stock trades below its Graham Number, you are paying less than the conservative valuation ceiling. The gap between price and Graham Number is your margin of safety – the cushion that protects you if earnings decline, if the market sours, or if your analysis contains errors. The wider the gap, the wider the protection. Buying a stock above the Graham Number means you are paying more than Graham's conservative limit. This verdict does not mean the stock will fall – growth, momentum, or market enthusiasm might push it higher still. It means you have no margin of safety. You are depending on the future to be better than the present, which is speculation rather than investment.

Margin of safety is Graham's central concept. Every valuation involves uncertainty; a company's earnings might fall, markets might turn, your analysis might contain errors, anything that may amount to a drop in stock prices or company value. The margin of safety is the cushion that absorbs these shocks. When you pay significantly less than a conservative estimate of value, you create room for things to go wrong without suffering permanent loss. The future is unknowable. We can't predict it; we can only build protection against being wrong.

Here is a simple example using KCB to illustrate such a cushion in action. At the end of 2019, KCB had reported EPS of approximately Ksh. 7.85 and a book value per share of about Ksh. 40, giving a Graham Number of around Ksh. 84(KCB Group, n.d.-a). Suppose an investor had bought KCB at Ksh. 40 in early 2020, at roughly half the Graham Number. When COVID-19 struck, KCB's earnings fell 22%, from Ksh. 25.2 billion to Ksh. 19.6 billion(KCB Group, n.d.-b). The new EPS dropped to approximately Ksh. 6.10. But even with this significant decline, the recalculated Graham Number fell only to around Ksh. 78, still nearly double the purchase price. The investor who bought with a margin of safety absorbed the shock.

Safaricom in 2021 illustrates the opposite outcome. In August of that year, the stock reached an all-time high of Ksh. 45.25. At the time, Safaricom's EPS was approximately Ksh. 1.71 and its book value per share roughly Ksh. 4.50, giving a Graham Number of about Ksh. 8.30(Safaricom PLC, n.d.). Investors buying at the peak were paying more than five times the conservative valuation ceiling—there was no margin of safety whatsoever. When Safaricom's earnings growth stalled and foreign investors began exiting, the stock had nowhere to go but down. By late 2023, the price had fallen below Ksh. 12, a decline of over 70%. The company had not collapsed. M-Pesa still processed transactions. And customers were still making calls. But the investors who bought without a cushion suffered severe losses because they had paid a price that assumed perpetual optimism. The margin of safety is not an abstract concept – it is the difference between a temporary setback and a permanent loss of capital.

The discipline is straightforward. Calculate the Graham Number. Compare it to the current price. If the price is well below the Graham Number, the stocks merit further investigation. If the price exceeds the Graham Number, move on – there are always other opportunities.

The Limits of the Formula

The Graham Number is a screening tool, not a complete analysis. It works well for companies with stable earnings and substantial tangible assets like banks, insurers, manufacturers and utilities where book value reflects real economic worth and earnings are relatively predictable. For such business, the formula does exactly what it promises.

But not every company fits in this mold. The formula is inapplicable to companies with negative earnings, since you cannot take the square root of a negative number. It also tends to undervalue companies with few physical assets but high earning power such as software firms, consulting companies, and brands built on intellectual property because the book value understates what these businesses are actually worth. In addition, it says nothing about qualitative factors; a company might pass the graham valuation test while facing declining markets, obsolete products, or incompetent management. The number tells you the price is cheap; it does not tell you the company is good.

These limitations are why the Graham Number is a starting point, not an ending point. It identifies candidates worth investigating, stocks where the price provides a margin of safety. The investor must then ask further questions. Is the company financially sound? Are earnings stable or growing? Does management act in shareholder's interests? The Graham Number gets you to the right neighborhood, due diligence tells you which house to buy.

The Practical Discipline

For any stock at the NSE, the process is the same. Find the latest EPS and BVPS from the company's annual report or financial data site. Plug the numbers into the formula. Compare the result to the current market price.

If you do this exercise across the market, patterns emerge. Some stocks trade at a fraction of their Graham Number – these are the candidates worth investigating. Others trade at multiples of their Graham Number – these are the stocks where enthusiasm has outrun fundamentals. The Graham Number does not guarantee profits, but it ensures you are never paying more than a conservative valuation would support.

In the final article of this series, we will examine three NSE stocks that crashed spectacularly like Safaricom, Uchumi, and Kenya Airways, and ask whether the Graham Number would have warned investors away before the fall. The answer, in each case, is instructive.

This is the fourth article in a series on value investing for the Nairobi Securities Exchange. The final article will apply Graham's principles to historical NSE crashes and examine what the numbers revealed before prices collapsed.

References

Disclaimer: This article is for educational purposes only and does not constitute financial advice. Investing in securities involves risk, including the possible loss of principal. Past performance does not guarantee future results. Readers should conduct their own research and consult with a qualified financial advisor before making investment decisions.