During the past few years, a plethora of headlines have proclaimed one “record fine” after another against the major banks at the heart of the 2008 financial crisis. After each settlement, Attorney General Eric Holder took a victory lap to show that, in America at least, no corporate misdeed would go unpunished.

But each of those compromises — from Citigroup’s $7 billion to JPM Chase’s $13 billion to Bank of America’s $16.65 billion settlements — left the public feeling increasingly shafted, not victorious. The question became not only whether certain offenders are too big to fail, but also whether they are too big, too complex, too powerful, and too expensively “lawyered up” to jail.

In Too Big to Jail, University of Virginia law professor Brandon L. Garrett explores this question with the sharp mind and attention to detail that exemplify his profession’s most positive attributes. Examining the intricacies of key federal cases and corporate bargains since the turn of the 21st century, Garrett considers Enron enabler and now nearly defunct accounting firm Arthur Andersen, international bribery convict Siemens, tax-shelter manufacturer KPMG, recidivist environmental and safety offender BP, drug-money-laundering colluder HSBC, tax-evader strategists Credit Suisse and BNP, and the Big Six U.S. banks at the heart of the recent financial crisis. Garrett paints a picture more disturbing than mere skepticism about the kid-gloves treatment these corporate villains have received. He portrays a justice system more likely to overlook a mega-billion-dollar crime (in return for tepid promises of reform) than a minor drug offense by a teenager.

Garrett characterizes corporate settlements as a game played between Goliath-sized corporations and David-sized prosecutors. The results aren’t Biblical. Rather, they are all-too-modern victories for the Goliaths. Fines are light when companies must pay them at all. When required as part of a settlement, even modest reforms are mostly ineffective or impossible to measure. Corporations can’t go to jail for their crimes, but lately, neither do their CEOs, who use techniques like the “ostrich” to bury their heads in the proverbial sand and are thus “blinded” to the misconduct of their employees. Notable exceptions to the successful ostrich defense include WorldCom’s former CEO Bernie Ebbers, who chose a trial over a settlement; Enron CEO Ken Lay and CFO Jeff Skilling; and Tyco’s Dennis Kozlowski. All of these men were convicted and sentenced in the early 2000s, though Lay died before serving prison time.

Garrett’s comprehensive and evidence-driven analysis probes widely held assumptions that federal prosecutors methodically coddle corporations and their executives with agreements that are not major deterrents against future crime. Often, he writes, they settle with these corporations because it’s better than nothing. However, to him, if there’s enough evidence to bring the cases to begin with, harsher penalties should be the result. In the past decade, as the number of cases against publicly traded companies has mounted, prosecutors have gone from extracting adequate punishment — recall the savings-and-loan crisis, when hundreds went to jail — to reforming companies as part of sweetheart deals. In the wake of this shift, Garrett questions whether prosecutors should or even can reform corporations, and how, given the rampant lack of corporate transparency, this effort could be practically tested. He concludes that if future settlements include reforms, they should be monitored much more effectively.

Another way prosecutors are increasingly letting companies slide is through deferred prosecution deals. If a company reforms itself by firing the relevant employees or implementing enhanced reporting or compliance mechanisms, it can soften its punishment or avoid a criminal conviction entirely. Another escape mechanism is the non-prosecution agreement, which is as slippery as it sounds. As Garrett says, such agreements “are far too important to avoid judicial review entirely.” Companies can also settle for fines, but even when they include restitution to victims, as in the case of the $1.7 billion JPM Chase forfeiture payment in the settlement surrounding Bernie Madoff’s Ponzi scheme, restitution doesn’t begin to equal the magnitude of the victims’ losses.

Garrett holds a soft spot for prosecutors and opts to address the larger issue of systemic weaknesses. His solutions include requiring prosecutors to pursue convictions except where collateral damage — for example, to innocent employees and shareholders — would be too severe. He would require judges to supervise deferred prosecution agreements, as too often they are not a part of the scrutinizing equation, though this may burden already overworked judges who may be unfamiliar with complex derivatives or pipeline safety measures. He suggests more detailed structural reforms when they are warranted; too often, settlements require none, or ill-conceived ones. He wants fines high enough to serve as deterrents, and harsher sentencing guidelines for repeat corporate offenders. Lastly, he believes the public and shareholders should know more about corporate crime. Prosecution agreements, detailed accountings of how fines are calculated, and progress reports describing compliance should be publicly available.

Garrett considers federal regulators as also contributing to the systemic problems and related recidivism arising from weak penalties. In July 2010, for instance, the SEC settled with Goldman Sachs on a $550 million fine for fraud charges related to one of its collateralized debt obligations. It also required Goldman Sachs to “reform its business practices” without eliciting any admission of wrongdoing. Six months later, Goldman completed an internal review of its policies and found only limited changes were necessary. The announcement preceded a series of lawsuits and settlements with five different regulators and a state attorney general. The charges included breaking securities laws, defrauding pension fund investors, and foreclosure abuses. Those settlements cost Goldman Sachs more than $4 billion, a fraction of the firm’s assets.

Garrett concludes that the problem of inadequate settlements usually comes down to size and might. The firms that receive the lightest federal treatment have grown so big — often because regulators approve their mergers — that holding them accountable for their crimes opens significant issues of collateral damage. And according to the federal government, because Wall Street banks are perceived to be so economically or systemically important to the country — a mantra propagated by Wall Street and accepted by the Washington elite — it is therefore risky to prosecute these firms too fervently.

This fear-steeped narrative is a critical piece of the legal compromise puzzle. How could the Department of Justice punish the very firms the government is subsidizing and protecting? This bias routinely benefits corporations at the expense of their victims. The CEOs go on to bigger bonuses or golden parachutes while the firms rack up new violations with impunity.

Garrett’s masterful book is as important as it is timely. Though he occasionally delves into legal history and jargon, he does so to deliver a more complete picture of indulgent settlements. Part of the rationale for weak settlements stems from various interpretations of the definitions of corporations and their constitutional rights, which he explains in detail. He balances this exposition with the specific elements of recent settlements, which usually leave the biggest corporations strong enough to commit new crimes. In the case of Siemens, the sole company that comes off as “reformed,” Garrett also discusses the positive role that corporate monitors can play to ensure compliance, though they often receive multi-million-dollar contracts in the process.

Garrett’s tone remains level throughout the book, as he lets the substantial facts and his legal analysis speak for themselves. As a result, his proposals for reform — stricter prosecution agreements, stronger and ongoing judicial review of compliance requirements, and greater corporate transparency — come off as evenhanded and eminently logical. Readers are left to experience their own emotions (mine tended toward anger) at the overly collaborative settlement process and whether there’s any hope for its redemption. I remain skeptical, especially after reading Garrett’s book.

We must be grateful to Garrett for compiling the most extensive database of corporate settlement information available today — the federal government’s own compilation is but a subset of Garrett’s — and for shining a light not only on the manner in which corporations skirt the law, but also on what must be done to curb them. Too Big to Jail is a cogent, exhaustively researched plea for saner and more equitable legal oversight. Garrett ends by noting that corporate crimes can overwhelm the limited resources of the justice system, but also that corporate prosecutions are themselves too big to fail.

Nomi Prins is the author of All the Presidents’ Bankers: The Hidden Alliances that Drive American Power. Her other books include It Takes a Pillage: An Epic Tale of Power, Deceit, and Untold Trillions (2009) and Other People’s Money: The Corporate Mugging of America (2004). She was a managing director at Goldman Sachs, a senior managing director at Bear Stearns, a strategist at Lehman Brothers, and an analyst at the Chase Manhattan Bank. She is currently a senior fellow at Demos, the public policy think tank.