Although there are considerably more than 800 publicly traded U.S. banks, many investors shy away from this sector. The reported financial statements look funny, and the results don't neatly fit into the handy-dandy screens and formulae that many investors like to use. That's a shame, because as Warren Buffett has shown in the past, well-chosen bank stocks can be as lucrative as industrial, health-care, or technology stocks.

For this column, I am grouping institutions of all stripes -- be they money-center banks like Citigroup (NYSE:C), regional banks like BB&T (NYSE:BBT), or savings and loans like Washington Mutual (NYSE:WM), under the generic title of "bank." While there are differences among the types, that's a topic for another day.

The basics
Let's start off with a brief examination of what banks do. Simply put, banks buy and sell money. Banks purchase money through deposits like checking and savings accounts; CDs; and debt or equity financing, or both. Banks sell money by making loans or investing in securities, or both. Buy low and sell high -- value creation at its finest.

Most, if not all, banks also make money from a constellation of fees and service charges. Some will even sell off almost all of their loans shortly after origination. Furthermore, many banks also have other income-producing operations, like asset-management and insurance businesses.

Interest rates are important to banks because they essentially determine the price at which money is bought and sold. But it's not particularly Foolish to make sweeping statements like "rising/falling rates are bad." Good bank managers are neither stupid nor helpless, and they can adjust to different environments.

To a large extent, then, the more important factor is how interest rates behave relative to a bank's management's expectations. If managers are preparing for higher rates, and those higher rates don't materialize, it's likely that the business won't perform as well as hoped (and vice versa with falling rates). Unfortunately, there really isn't a handy ratio or formula to easily establish management's expectations. It requires paying attention to Securities and Exchange Commission filings like 10-Ks and 10-Qs and management comments during conference calls.

There are many moving parts to a bank. It doesn't really have revenue in the normal sense of the word, but you can look at the combination of interest income (minus loss provisions) and non-interestincome as a proxy. Clearly, you'd like to see these numbers going up.

Deposit growth and loan growth are also both important. If deposit growth is weak, other more expensive sources of funds might have to be tapped. Likewise, if loan growth is sluggish, it will be more difficult for the bank to earn a profitable spread on the money it controls.

Let's also not overlook earnings growth -- a growing bottom line is just as important to a bank as any other company. So, for all of the fancy or confusing concepts that follow, don't forget that income and per-share earnings growth are still important.

As is the case with any business, profitability is paramount to banking success. Strong profitability can easily spell the difference between an average stock and an above-average winner.

Net interest margin is a good baseline measurement of the profitability of a bank's core lending and borrowing business. Net interest margin is basically the difference between interest income (loans, securities, et. al.) and interest expense (deposits, borrowed funds, et. al.). In some ways, I suppose you could think of it as a bank's gross margin.

When analyzing net interest margin, I'd suggest you do so in the context of similar banks and their strategies. Among banks in similar lines of business, higher margins can be a sign of great management. But it could the result of riskier lending policies. Narrower margins can suggest trouble on the deposit side and a higher cost of funds. Or it could mean more conservative lending practices. Context matters when comparing numbers.

The efficiency ratio is another useful metric. There are different ways to calculate this number, but the fundamental underpinning is the same: It is designed to measure the ratio of non-interest expense to income. Ideally, you want this number to be as small as possible -- the smaller the number, the less being spent on things like marketing, salaries, and branch expenses, and the better the earnings.

A bank's capital structure, asset quality, and liquidity are all important. Because the spreads on borrowing and lending money are relatively narrow, banks employ considerable leverage to pump up their earnings power. To make sure that banks don't overdo it and imperil their solvency, bank regulators impose limits.

To be "well capitalized," a bank must have a ratio, expressed as a percentage, of risk-based total capitalto total assets of 10%. ("Risk-based" refers to certain adjustments that are made to reflect that some assets -- like cash and treasury securities -- are less risky than others.) Likewise, shareholders can examine the ratio of shareholders' equity to assets to get a grasp on whether the company is increasing or decreasing its leverage.

Asset quality is important; if people don't repay their loans, the bank in question will have some problems. Numbers to look out for include net charge-offs, the percentage ofallowance of loan losses to total loans, and the percentage of non-performing loans tototal loans. Watch their trends to see whether asset quality is going downhill.

Liquidity is also important in the banking world; it basically reflects how easily a bank can lend money. If a bank is very liquid and has a lot of cash on hand, it can respond very quickly to new opportunities. Of course, there is an opportunity cost associated with keeping cash on hand -- you may lose out on what you could make by lending it out. So if a bank is too liquid, it's leaving money on the table. There are two quick ways to check liquidity -- the percentage of assets held in cash and treasury securities and the loan/deposit ratio. Ideally, this latter number should be near 100% for most well-capitalized banks.

All other things being equal, it's generally a good idea to invest with the folks who've proved that they can make the most money with a given amount of capital. One way to make that judgment is to look at a company's return on assets and return on equity. In a sense, these metrics capture the end result of a management's decisions on liquidity, leverage, credit quality, and other operational options.

The one note of caution I'd sound on return on assets is that you should make sure you're comparing banks with similar business models. Banks that generate a lot of non-interest income, like Bank of New York (NYSE:BK), can sometimes require less assets on a dollar-for-dollar basis than more lending-focused banks. So comparing across these lines can give you misleading results.

Return on equity is also highly useful for me. I like that you can break it down into net margin, asset turnover, and financial leverage and study how and why one bank may be better than another. Moreover, I've found that banks where management has consistently produced above-average return on equity are often above-average performers in the stock market.

Know the business
There is no such creature as a "typical bank" -- each has its own nuances and individual risk factors. Take beleaguered DoralFinancial (NYSE:DRL), for instance: While it looked good by the numbers and had a heck of a run, there was a ticking time bomb in the way the company valued aspects of its mortgage operations. Had you simply relied on the numbers and not dug into the 10-K, you could have easily been fooled into thinking the bank was safer than it really was.

Knowing the business is also important when you compare banks. Comparing a huge money-center bank like KeyCorp (NYSE:KEY) with a regional bank like Bank of theOzarks (NASDAQ:OZRK) just doesn't make sense. So the more you know about the business, the more likely you are to understand what the numbers are really telling you.

Don't overthink this
We have offered this brief treatment on bank stock analysis to help you with your evaluations of these stocks. But as is always the case, numbers tell only a certain side of the story. It's important to do your own due diligence and strive to really understand the policies, strategies, and expectations of the company in question before buying a small piece of it.

True, the banking sector has some unusual ratios and metrics for measuring performance, but don't get bogged down in the details. Ultimately, the pattern of success behind a bank stock investment is little different than that for most any other company -- a pattern of consistent growth in earnings, dividends, and assets, skillful and honest leadership, and expectations of further growth.

This article originally ran on August 31, 2005. It has been updated.

Are dividends your thing? Join Mathew Emmert in his quest to find the best dividend-paying stocks around. A free trial to Motley Fool Income Investor is available today.

Fool contributor Stephen Simpson owns shares of BB&T, but has no financial interest in any other stocks mentioned (that means he's neither long nor short the shares). The Motley Fool is investors writing for investors.