How a mid-level tip ring toppled Wall Street's junk-bond king

1

Levine's ordinary tip market

Stewart shows the scandal began with a mid-level banker turning routine office access and Bahamian secrecy into a private information market, not a criminal genius.

Dennis Levine worked mergers desks at Citibank, Smith Barney, Lehman Brothers, and finally Drexel Burnham Lambert, always near confidential deal flow. Around 1980 he started trading on that knowledge through an account at Bank Leu in the Bahamas, chosen because local secrecy laws made ownership nearly impossible for U.S. regulators to trace.

He recruited a small circle of fellow bankers and lawyers—Robert Wilkis, Ira Sokolow, and David Brown—who fed him tips from their own firms for a share of the profits. The setup thrived because Wall Street M&A ran on advance knowledge of takeovers, and the people holding that knowledge were young, ambitious, and underpaid relative to the fortunes forming around them.

By the mid-1980s Levine was moving millions through the Bahamian accounts ahead of one announced deal after another. Stewart's point is that scale came from an ordinary practice—treating deal information as a personal asset—that had simply never faced serious investigation. Levine's network is the seed from which the rest of the cascade grows.

In practice: His pattern of winning bets ahead of successive merger announcements became statistically impossible to explain as luck, which is what finally made the Bahamian accounts impossible to ignore.

2

Boesky's blurred edge

In Stewart's account, Boesky's legal arbitrage business became a channel for illegal tips because payments for research could hide purchases of advance knowledge.

Ivan Boesky ran one of Wall Street's most prominent risk-arbitrage operations, betting on which companies would be acquired and at what price once deals were rumored or announced. The business was legal and demanded real skill to judge which transactions would close. He cultivated a public image as a brilliant, disciplined risk-taker who told business students that greed, properly channeled, served the market.

Behind that image he paid for edges analysis alone could not supply. Money sent to Dennis Levine as consulting or finder's fees bought knowledge of deals before they became public. Because arbitrageurs routinely paid for research and kept banker relationships, those payments could sit inside normal-looking arrangements.

Scale magnified every edge: the firm managed billions, so modest informational advantages on major deals became enormous profits. People also moved between firms and roles faster than internal walls separating bankers from traders could be rebuilt, leaving employers blind to the partnership for years.

In practice: Fees labeled as consulting or finder's payments let Boesky buy Levine's advance deal knowledge inside arrangements that resembled ordinary research relationships.

3

Junk bonds rewrite takeovers

Milken's high-yield desk, in Stewart's telling, proved that quick access to speculative capital—not board approval or comparable size—could decide who acquired a company.

Before Milken, corporate raiders had limited means to finance hostile bids for companies many times their own size. From Drexel's Beverly Hills trading desk, nicknamed the X-ready desk, he built a market for high-yield, high-risk bonds that could be issued quickly to fund those bids.

His annual Predators' Ball in Beverly Hills became the industry's central gathering, drawing raiders including Carl Icahn, Nelson Peltz, and Ronald Perelman. They used the bond-buying network to chase outsized targets. Personal pay matched the machine: in 1987 alone Milken received roughly $550 million.

Stewart argues the financing machine mattered beyond any single deal because it rewired board incentives nationwide. A company no longer needed a friendly peer acquirer; anyone with access to Milken's network could bid, pressing boards to keep share prices high or face a fight. Defenders called this shareholder discipline; critics said it loaded firms with debt and short-term thinking. The book does not settle that economic fight, but treats the operation itself as the prize the government eventually pursued.

In practice: Ronald Perelman used Milken's junk-bond financing capacity to acquire Revlon in 1985, a target far larger than conventional means would have allowed.

4

Siegel's cash deliveries

Siegel's secret cash-for-tips arrangement with Boesky showed that seniority and reputation offered no real barrier once large money appeared.

Martin Siegel ranked among Wall Street's most respected merger advisers, first at Kidder Peabody and later at Drexel, trusted by corporate clients with their most sensitive acquisition plans. Starting in the early 1980s he quietly supplied Ivan Boesky with advance information on pending deals.

Payment arrived as physical cash through intermediaries at prearranged locations, including sums reportedly left in a briefcase. Over time those payments reportedly totaled around $700,000. Unlike Levine, Siegel was not a junior employee hunting a shortcut; he was a senior partner whose judgment firms and clients prized precisely because he seemed above such bargains.

His participation demonstrated that the informational advantages of top M&A work created temptation regardless of rank, and that industry self-policing—built on trust and standing—had no mechanism to catch someone at his level. Once Boesky cooperated, Siegel became a second independent thread spanning multiple major firms. He pleaded guilty and received a lenient sentence for cooperation, but his fall showed the problem was not confined to one firm or one employee type.

In practice: Cash for Siegel's tips moved through intermediaries at prearranged spots, including a payment reportedly left in a briefcase.

5

Unraveling by persistence

Stewart presents the case's opening as slow pattern detection and pressure on a Bahamian bank, showing concealment still fails when regulators refuse to quit.

The investigation that eventually reached Milken began almost mundanely. Merrill Lynch received an anonymous letter, apparently from a disgruntled trader, alleging suspicious trading tied to Bank Leu accounts in the Bahamas. Gary Lynch's SEC enforcement division had already been tracking unusually well-timed trades ahead of merger announcements; the letter gave them a concrete institution to pursue.

Bahamian law protected account-holder identities, so the SEC could not simply demand names. Prosecutors combined diplomatic pressure with legal threats against the bank itself until authorities allowed limited disclosure. That path led to Dennis Levine, arrested in May 1986.

Stewart uses the sequence to stress the limits of clever concealment: Levine's offshore structure was sophisticated by ordinary standards, yet it assumed regulators would abandon a jurisdictional dead end. They did not. Confronted with the evidence, Levine cooperated rather than face a long prison term, named his recruits, and critically identified Boesky as a buyer beyond the junior circle. One unglamorous arrest became the hinge that opened everyone above him.

In practice: Sustained pressure on Bank Leu produced limited Bahamian disclosure that pointed to Levine; his May 1986 arrest and cooperation named Boesky.

6

Boesky wears the wire

Boesky's secret recordings after a record penalty turned one defendant's leverage into direct evidence against suppliers far harder to reach.

Facing Levine's testimony, Ivan Boesky negotiated a November 1986 plea requiring $100 million in fines and disgorgement—then the largest such SEC penalty—and a guilty plea to a single securities violation. In a controversial arrangement, prosecutors let him continue trading and appearing publicly for a period after the deal but before announcement, allowing positions to be unwound before the news moved markets.

More important to the narrative, Boesky spent months secretly recording conversations with people who had supplied him information, including Martin Siegel. The tapes gave prosecutors direct evidence rather than secondhand accusation, material far harder for defense lawyers to dismiss.

Stewart treats the cooperation as the investigation's true turning point because Boesky's relationships reached into Drexel and toward Milken through alleged stock-parking arrangements in which Milken's desk held shares to disguise ownership and manipulate reported positions. Whether Boesky's account of those arrangements was fully reliable stayed disputed: he had every incentive to please prosecutors and shrink his own sentence. He ultimately served less than two years. The book presents the cooperation as both indispensable and inherently compromised by self-interest.

In practice: After the plea, Boesky secretly recorded months of talks with suppliers including Siegel, producing tapes prosecutors could offer as direct evidence.

7

Drexel under the RICO threat

Giuliani's readiness to seek a RICO indictment that could freeze assets before trial forced Drexel's guilty plea, substituting leverage for a verdict on the charges.

By 1988 federal prosecutors under U.S. Attorney Rudolph Giuliani had assembled a case against Drexel built substantially on Boesky's cooperation and internal firm records. Rather than proceed straight to trial, they made clear they were prepared to indict under the Racketeer Influenced and Corrupt Organizations Act, a statute originally aimed at organized crime.

A RICO indictment carried power to force pretrial asset freezes. For a securities firm dependent on continuous capital access and counterparty trust, that freeze would likely mean immediate collapse no matter how a later trial ended. Facing the threat, Drexel's board pleaded guilty in December 1988 to six felony counts, paid $650 million in fines and restitution, and agreed to cooperate against Milken, opening files and personnel.

Stewart presents this as pivotal and troubling. By its own later account, Drexel's leadership believed the firm could likely defend the specific charges at trial, yet concluded the RICO risk made survival to trial impossible regardless of guilt. The plea therefore proved little about the truth of the allegations and everything about the firm's inability to absorb indictment risk. That cooperation is what stopped Drexel from shielding Milken and supplied evidence that had leaned heavily on secondhand testimony.

In practice: Drexel pleaded guilty to six felony counts and paid $650 million after prosecutors signaled a RICO indictment could freeze assets before any verdict.

8

Milken's narrower landing

Milken's six-count plea, driven by trial risk and exhaustion, covered stock parking and tax conduct more than the junk-bond takeover machine tied to his public image.

Milken was indicted in March 1989 on 98 counts covering racketeering, securities fraud, and related charges—a sprawling case built from Boesky's testimony, Drexel's cooperation, and years of financial records. A trial could have lasted months; racketeering convictions carried potential sentences measured in decades.

In April 1990 he agreed to plead guilty to six lesser felony counts, primarily stock-parking arrangements with Boesky and tax-related conduct, rather than charges that directly attacked the high-yield financing business that had made him famous. He paid $600 million in fines and restitution and received ten years in prison, later reduced to roughly two after cooperation and good behavior.

Drexel, already weakened by the 1988 settlement and a shifting bond market, filed for bankruptcy in February 1990, eliminating roughly 5,000 jobs. Stewart frames the ending carefully: the plea resolved the case without adjudicating in open court whether high-yield takeover financing itself was criminal or simply aggressive and unpopular. The admitted counts were narrower and more technical than the public image of the prosecution, a gap the book leaves visibly unresolved.

In practice: After a 98-count indictment, Milken pleaded to six counts focused on stock parking and tax conduct rather than a wholesale attack on junk-bond takeover finance.

9

Incentives over bad apples

Stewart argues Levine, Boesky, Siegel, and Milken expressed pay structures, paper Chinese walls, and a culture that treated self-interest as professional virtue.

Stepping back from the individual cases, Stewart treats the four men not as aberrations inside a clean industry but as extreme expressions of 1980s Wall Street incentives. Compensation rewarded results, not the means used to get them. Firms competed fiercely to sit near the biggest deals and therefore tolerated aggressive behavior from the people closest to those deals.

Internal compliance systems—the so-called Chinese walls separating deal teams from trading desks—existed mostly on paper. Enforcement depended on individual discretion, and people moved between roles and firms faster than any wall could be rebuilt. Business schools, financial media, and Wall Street itself celebrated aggressive risk-taking and personal wealth as markers of success; figures like Boesky were publicly praised for embracing self-interest as a legitimate creed.

That climate made the boundary between celebrated ambition and criminal behavior harder for participants to see in real time, even though the legal line in most of these cases was not actually ambiguous. Stewart does not claim every professional broke the law. He argues the specific mix of incentive, secrecy, and self-regulation made a scandal of this scope more likely than not, and that punishing a handful of people did not by itself rewrite those structural incentives.

In practice: Chinese walls meant to keep bankers' deal secrets from traders depended on personal discretion and failed when people changed firms or roles faster than the walls could be rebuilt.