Few people now remember that the Internal Revenue Service once prepared tax returns for free. Tired of chronic complaints about its many mistakes, the IRS began discontinuing this service in the mid-1950s. News of this decision arrived in Kansas City at the same time that Henry Bloch's family bookkeeping business also planned to abandon tax preparation. But one of Bloch's clients had a better idea. He convinced the bookkeeper to capitalize on the government's retreat by offering to do anyone's taxes for a simple, five-dollar fee. Bloch's firm was promptly deluged with business.

In 1955, Henry and Richard Bloch changed the spelling of their firm's name to H&R Block to avoid confusion. The company grew quickly, becoming the national leader in tax preparation and currently completing about 20 million US tax returns annually. In the 1960s, Block started training legions of temporary tax preparers in company tax courses, now taken annually by some quarter-million Americans at a cost of up to $199 per class.

To hear Block tell its story, the firm embodies the democratic spirit — born from offering small businesses and individuals of modest means the tax and financial planning services traditionally available only to the wealthy. But how well does Block's reality match its hype? As a temporarily underemployed industry analyst with basic accounting knowledge, circumstance led me to find out. In October 2006, I enrolled in a Block tax course at its Pleasant Hill office. My classmates included a lawyer, more than one accountant, and a couple of financial counselors.

As I researched my potential employer, I naturally wanted to find out what I would be paid as a Block tax preparer. You would have thought I was asking the Pentagon how to steal hydrogen bombs. I tried calling a district office and even national headquarters, but the consistent answer was silence. For most new employees and many more, the truth was that Block paid something virtually indistinguishable from California's minimum wage. Although there was a bonus system, relatively new employees barely benefitted from it, if at all. The company's unresponsiveness on this issue seemed to highlight a central tenet of its business strategy.

Obfuscation about compensation was a standing joke throughout our 66-hour course. In response to persistent questions from me and some of my sixteen classmates, our teacher, who I will call Val, claimed that she had no idea what Block paid since she hadn't prepared taxes in years. That didn't stop her from regularly regaling us with the seductive Block dream. "You can make any amount of money you want to," she allowed. Val periodically recited the tale of "one woman" she had "heard about" who made $40,000 each three-to-four-month tax season, supporting herself for the year.

Doubtless there were experienced preparers who did relatively well; tales of their success were circulated relentlessly. But a very large proportion of tax preparers were in their first few years.

At our next-to-last class session, we were honored with a visit from the company's Walnut Creek district manager, who I will call Jimmy. He casually informed us for the first time that we would "actually" need to take another course before we could work for Block. It would teach us the company's automated tax-preparation system, which we would generously be paid the minimum wage to learn.

However, Jimmy did provide what I thought was our first hint of what we might expect to earn at Block. "How would you like to make your living in four months of the year?" he asked, echoing Val. Only in our first year would our up-front salary have to be relatively basic, he told us. The way to earn much more was through the company's legendary, production-based "bonus," which also set the level of our base pay in subsequent years. For example, Jimmy explained that if we made $20 an hour in salary and bonuses as first-year tax preparers, when we came back for our second year our base pay would automatically rise to 80 percent of that. Finally, we had the clue we'd been seeking: a first-year tax preparer might make $20 an hour with bonus. This was extremely disappointing, but presumably only an average.

Imagine my surprise when Jimmy told me the next evening that "realistically" no first-year tax preparer in his district made more than $17 an hour the year before — and even that was atypical. Second-year employees did little better. Jimmy said we needed to have "reasonable expectations," adding, "Maybe this isn't something you really want to do." After all, he said, most new employees "weren't in it for the money" the first few years, but rather for "the experience."

"We just don't pay people $25 an hour," he added, as if that were the pinnacle of economic achievement no-accounts like me hoped to squeeze out of unsuspecting multinational corporations.

Jimmy certainly seemed to have no worries about my reaction. Given the way Block churns out tax preparers, he had far more candidates than he could ever use.

Part of what ensured that Block had too many candidates already seemed clear to me. Val had told us early on that if we didn't know the answer to one of the tax course's many tests we should just check with her, and she would tell us.

"What I want is to make sure you all know how to do taxes, not to get a score on a test," she said, implying that studying to earn individual grades contradicted this higher goal. I asked explicitly, both in class and privately, if this approach was endorsed by the company. Val, who said she had taught at Block for many years and worked as a manager and tax preparer for many more, assured me it was "the Block way." When Jimmy confidently predicted that we would all pass the final and class with flying colors, we thought we understood how he knew.

Most of our time in class was spent mechanically reviewing problems assigned the night before. Val displayed the correct answers on the projector, taking little apparent interest in whatever mental process led to them. Students looked at the official answers until they or someone else figured out the problem. Whenever someone raised a hard question or asked about a complicated issue, Val would say irritably: "You'll never see a case like that." She declined to address complexities or potential contradictions — which, of course, abound in tax law. We should look it up.

Our frequent quizzes were done first on a practice basis, with students advised of the right answers before final grading. The midterm was done in teams, with Val again providing correct answers before grading. On the all-important final, which accounted for a majority of our grade, Val told us to ask if we had any doubt about our answers. Together, we reviewed the answers aloud before grading, like much younger students.

This policy ensured that everyone passed, but it also seemed to render meaningless the California licensing requirement designed to ensure that certified preparers know tax law.

There was no way to assess the expertise of anyone in class from grades. Nor was there any detectable effort to assess skills and abilities in the hiring process I underwent. But given how little we actually had to know in our jobs, this approach probably made sense for Block.

The Block course did teach a lot about taxes, at least for those who did the homework. I'd still recommend it to anyone who wants to know how to prepare different kinds of taxes. But to work for H&R Block, did it matter how much we knew? There seemed to be little connection between the state-mandated course content and the reality of the office in which I worked.