At the beginning of Warren Buffett's annual letters to the shareholders of Berkshire Hathaway (NYSE:BRK-A) (NYSE:BRK-B), he always starts with a comparison of how it performed versus the S&P 500. In the most recent letter, for instance, he showed that, since 1965, Berkshire Hathaway's book value per share has increased by a total of 693,518% compared to the S&P 500, which has grown by a cumulative 9,841%.

Both are good, but suffice it to say that Buffett's track record is significantly better.

Fair enough, but here's the question: Why didn't Buffett compare Berkshire's share price to the S&P 500? Wouldn't that be a more appropriate comparison?

The answer is both yes and no. It's "yes" because by looking at share price, at least on a superficial level, you're comparing apples to apples. However, it's "no" because Buffett doesn't have the same level of control over the price of Berkshire's shares on any particular day that he does on its book value over the long term.

To put it slightly differently, book value per share exhibits much less noise than does share price.

This is a valuable insight that applies with equal force to companies and industries beyond Berkshire Hathaway. For instance, as you may have guessed from the title, I think this same approach should be used by investors in bank stocks.

When you look at Wells Fargo (NYSE:WFC), for example, it's difficult -- if not impossible -- to even remotely predict how quickly its share price is bound to increase during the years to come. The obvious way is to look at the compound annual growth rate of its share price in the past. But the problem with this approach is that it captures the noise from investor sentiment, which causes a bank's price-to-book value ratio to fluctuate widely at different stages of the credit cycle.

By contrast, if you focus exclusively on book value per share, the picture becomes less complicated. As you can see in the chart below, absent the aberration during the financial crisis, Wells Fargo's book value per share has increased at a relatively consistent pace over the past decade.

WFC Book Value (Per Share) Chart

Thus, by focusing on book value, investors can get a better, albeit still imperfect, sense of what to expect in the future.

For example, let's assume that a bank generates an average annual return on equity of 12%. And let's also assume that it distributes a third of this to shareholders, meaning that it retains only two-thirds of its earnings. It follows that our hypothetical bank would be expected to increase its book value per share by an average of 8% a year.

Now, of course, there are a number of complicating factors. Among others, it's generally necessary to factor in net share buybacks -- that is, the difference between how many shares a bank issues each year to its employees and how many shares it buys back as a part of its capital allocation strategy.

Additionally, as I alluded to earlier, the increase in a bank's book value doesn't translate directly into an increase in share price -- which is ultimately how investors monetize their stake in a company. Sometimes, market sentiment overvalues banks; sometimes, it undervalues banks.

Over time, however, these factors offset. As Benjamin Graham has famously noted: "In the short run, the market is a voting machine, but in the long run it is a weighing machine."

The point here is that, like Buffett, investors would be wise to pay less attention to a bank's historical share price performance, and more to its record of increases or decreases in book value per share. Not only does this give you a potential leg up in gauging future returns, but it also allows you to more accurately assess the fundamental performance of the bank itself.