Published by Harvard Business School Press, part of the "Leadership for the Common Good Series", of which I have previously read Bazerman and Watkins _Predictable Surprises_, however, which I do not appear to have reviewed in this blog. I got rid of Bazerman and Watkins' book in the Really Big Purge within the last year, because I felt like it was yet another too-superficial and often misleading treatment of some of Kahneman and Tversky's work, unless I have confused it with another book, but never mind that now.
Recently, the IRS has been attacked for allegedly politically motivated differential treatment of certain non-profits applying for 501(C)4 status. As a fan of Colbert, I was a little confused why any of them applied anyway, as it was not clear they had a filing requirement. Also, given that the groups in question oppose taxation/more taxation/modification of taxation other than net reduction overall AND reduction of their own and so on and so forth, I perceived any targeted attacks by the IRS as not partisan in motivation but rather what one hopes for in a competent tax collection authority (that is, if you're going to have tax collectors, you would like them to be making sure people pay, and groups which are in opposition to payment are rationally their enemies, and costing honest taxpayers as well).
While I have been filing tax returns since, um, either 1985 or 1986 (I forget), but only filing complex tax returns comparatively recently (post 1996, call it), my experiences with the IRS have been primarily _after_ the most recent previous set of scandals touching upon the IRS and which are the subject of Rossotti's book. And my experiences with the IRS, while not extensive, weren't trivial either and were incredibly positive. Did I wind up paying more money? Yup: due to error, omission and bad judgment on my part, generally speaking if the IRS and I have not agreed on something, it involved me writing them another check. Which I really don't have a problem with, as long as I'm told how much the check should be for and it is applied to my account and so forth and so on and that is the end of that. Which is has been. I watched a little of the scandalation, but with incomprehension, because I actually kinda like the IRS and I don't much care for the people doing the complaining. But I figured I should do a little research to make sure I wasn't blowing off something important.
Alas! There are very few books written for a general audience about the IRS. This one seems to be the most recent, a memoir written about being IRS Commissioner from 1997-2002. Rossotti was about to retire from a consulting business that was accustomed to making sure monster projects stayed on/got back on track for Very Large Companies, so he was a logical pick for Rubin and Summers to install at the IRS in the wake of those 1996-7 scandals. Rossotti delayed the second Really Big Contract (15 years, design build, little clarity on who was responsible for failures/delays -- no wonder he was appalled!), brought in A Few Good Men and Women, promoted many more, re-orged the whole shebang and made sure the IRS survived Y2K. Then he started the requirements phase of the Really Big Contract (oh, and the Bush stimulus checks and 9/11 occurred during his tenure). He also redirected audit focus.
About the Reorg: according to Rossotti, the last big reorganization occurred in 1952 and was intended to professionalize the agency and put an end to political influence (they used to appoint people, top to bottom! Jeepers! Er ...). That worked, but the world in 1952 was a very different world. He mentions jet airplanes and long-distance phone calls being expensive; equally relevant, everything was paper based in 1952 and no one in their right mind moves paper a foot further than they have to. The IRS was organized _locally_, with loose policy direction at the national level. With the exception of the 1960s era creation of the master file, fed by the local organizations, all IT and other innovation occurred within these geographic fiefdoms. Rossotti reorganized it into 1990s era business units, with tight national control and nationally controlled procurement of IT, management of facilities and personnel and so forth. This reorganization also enabled him to dramatically flatten the hierarchy, reduce silo-ing and make it possible to develop technical expertise on various elements of tax law. It also let them set up real call management, thus reducing busy signals, reducing wait time on hold and dramatically reducing wrong answers and repeat phone calls that result from wrong (or no) answers.
Predictably, because the IRS is So Big and because their IT was So Old and because the tax code is So Complex, the requirements phase of the Really Big Contract was a *insert your preferred swears here*. Despite a delay being inserted because of known problems with the requirements, the first deliverable on replacing the master file was delayed because of inaccuracies in the requirements phase; the master file would not be replaced until 2004 or thereabouts. [ETA: Still working on this, apparently. http://en.wikipedia.org/wiki/Customer_Account_Data_Engine Also, batching has finally gotten down to daily! Woot! http://www.fiercegovernmentit.com/story/tigta-irs-must-improve-cade-2-requirements-management-testing-security/2013-01-03] If you were hoping for a book about bring the IRS up to our adult lifetime technology, this isn't it. I'm currently thinking about just going back and reading the IRS Data Books for the last, oh, 100 years because I suspect that's the best way to understand what happened and I can safely skim all the tables and stick to the explanatory sections. Also, all online! Woot! I am not, however, optimistic about how detailed the explanation of the upgrades will be. Still, I sampled 1960 and it was interesting. YMMV.
Audits until 1960 didn't really focus on lower income tax payers, but around that time, the agents who were capable of auditing more complex returns of higher income payers were re-deployed internally to more qualitative analysis (possibly in support of the requirements phase of the then new automation initiative that would create the master file?), and the IRS started doing more lower income audits. I'm not sure that bus was ever turned again, until Rossotti showed up and fixed the audit-focus to go after tax shelters. Which is scary and sad.
Rossotti concludes the book with some suggestions for simplifying the tax code. Given that the last big simplification we had (1986) resulted in a dramatic increase in inequality and a tax code which mushroomed back to Even More Complex in about a decade, I'm not convinced that simplification is such a brilliant idea. I do agree that the specific simplifications he advocates (making the definition of a child for tax purposes more consistent, reducing the complexity of tax/retirement account incentives) are probably worth doing. He notes that the IRS was distributing more program dollars than the SSA -- and at a lower _cost_ than the SSA, which led me to a very different set of conclusions than he came to (if it's really that cheap to administer benefits through the IRS, why bother to have any other program at all? Seriously!).
Rossotti is a real classic of the aging 1990s era business critter: he's a technocrat who finds lawyers infuriating and politicians (who make announcements with no basis in reality) misguided or worse. Of course I'm going to basically like him. But at the same time, I see a lot of limitations in that world view. If anyone knows of a more recent book about the IRS, I'd love to know about it. If anyone out there is thinking of writing one, I look forward to the results of your labours.
ETA: I bought this used on Amazon. I plan on keeping it, at least for a while.