On May 18, 1998, the Department of Justice filed an antitrust lawsuit against Microsoft, seeking ultimately to break up the company. The case was controversial at the time and remains controversial now, but I would also argue the case is misunderstood today, and the Microsoft of today is not the Microsoft of the 1990s, and the Microsoft Antitrust Case of 1998 had something to do with that.
It’s popular to say today that the Microsoft antitrust case of 1998 was unnecessary and the market figured out how to handle Microsoft. But provisions in the settlement kept Microsoft from behaving toward Google the way they behaved toward Netscape and Lotus.
Why the government brought the Microsoft antitrust case

In the suit, the U.S. government accused Microsoft of illegally monopolizing the web browser market for Windows, primarily through the legal and technical restrictions it put on the abilities of PC manufacturers (OEMs) and users to uninstall Internet Explorer and use other software such as Netscape and Java.
Having a monopoly isn’t illegal, but using one monopoly to get another monopoly is.
At the time, popular opinion was divided. Disliking Microsoft wasn’t as hip and chic in 1998 as it is today, but the Microsoft fanboys I knew in the 90s hated it when you called them one. The hilarious contradiction is that people did indeed line up at midnight to buy Windows 95 the same way people line up at midnight to buy Apple’s new product, but few wanted to admit to standing in line to buy Windows 95. Whereas people brag about standing in line at midnight to practice their Apple iDolatry.
What people thought of the Microsoft antitrust case in 1998
In a lot of cases people’s opinion on the Microsoft case said more about their politics than what they thought about Microsoft. I knew dozens of Apple fanboys who opposed the lawsuit because they were conservative. In their case, their politics was more important to them than what Microsoft would have done to Apple if left unchecked.
Most people I knew at the time opposed the case and thought Microsoft would win one way or another. Those who didn’t know much about the computer industry or Microsoft thought Microsoft would win outright. The ones who knew Microsoft theorized Microsoft would throw the case and win on appeal, because the appeals court tended to be more conservative.
Microsoft’s behavior at trial looked like someone not really trying to win. Bill Gates was evasive and combative, arguing over the definitions of simple words like “compete,” “concerned,” “ask,” and “we.” Microsoft provided falsified videotaped evidence as well. On November 5, 1999, Judge Thomas Penfield Jackson issued findings of fact, holding that Microsoft took actions to crush threats to its Windows monopoly, including applications from Apple, Java, Netscape, Lotus Software, RealNetworks, and others. On April 3, 2000, he issued his conclusions, holding that Microsoft violated sections 1 and 2 of the Sherman Antitrust Act.
How Microsoft ran out the clock on the DOJ
On June 7, 2000, the District Court ordered a breakup of Microsoft as its remedy, splitting its operating system business from the rest of its software business. Microsoft immediately appealed.
On June 28, 2001, the Circuit Court overturned the rulings, including the holding that Microsoft should be broken up, but did not overturn the findings of fact. The case went back to the District Court, this time under Judge Colleen Kollar-Kotelly.
But ultimately it didn’t matter. The Department of Justice started the case under President Bill Clinton, a Democrat and a professed liberal. In November 2000, George W. Bush, a Republican and a professed conservative, was elected president. Bush’s Justice Department was less interested in breaking up a monopoly than Clinton’s had been. On September 6, 2001, the Department of Justice announced it was no longer seeking to break up Microsoft and would seek a lesser penalty. Microsoft had run out the clock.
On November 1, 2001, the Department of Justice settled with Microsoft. It required Microsoft to share its APIs with third parties but did not require changing any code or prevent Microsoft from tying any other software to Windows in the future. Nine states and the District of Columbia, who had pursued the case along with the DOJ, argued it didn’t go far enough to curb Microsoft’s anti-competitive practices. But on June 30, 2004, the Circuit Court approved the settlement.
Microsoft’s obligations under the settlement expired November 12, 2007.
Was the Microsoft antitrust case necessary?
One popular narrative is that the legal action wasn’t necessary and the free market prevailed over Microsoft. After all, Apple made its big comeback, and Google Chrome became the dominant web browser, and both companies are as valuable or more valuable than Microsoft today.
How it helped Apple
But tying up Microsoft in court gave Steve Jobs time he needed to figure out what to do. Jobs figured out that the key to beating Microsoft was to play the infinite game against them, but that was something he had to learn. He had not played the infinite game in his first stint at Apple, at least not consistently, nor had he at NeXT.
Jobs also needed time to get OS X ready. Mac OS X did not reach the market until March 24, 2001. Apple wasn’t in a position to compete with Microsoft until that happened. So-called “Classic” Mac OS was obsolete and couldn’t compete with Windows 2000, let alone Windows XP. But OS X could.
The Department of Justice tying Microsoft up in court until November 2001 gave Apple the time it needed to catch up and produce a viable alternative operating system.
How it helped Google
Google finally ended Microsoft’s dominance over web browsers with Chrome. Google launched Chrome in 2008, when it was 10 years old. If Microsoft had been free to do whatever it wanted, would Google have survived to 2008? Microsoft first entered the search business in 1998, the same year Google launched. If Microsoft had felt free to do to Google what it did to Lotus, Netscape, and Real Networks, Google may not have made it to 2008.
Microsoft was watching its behavior after the settlement. Its 2008 Annual Report contained the following statement:
Lawsuits brought by the U.S. Department of Justice, 18 states, and the District of Columbia in two separate actions were resolved through a Consent Decree that took effect in 2001 and a Final Judgment entered in 2002. These proceedings imposed various constraints on our Windows operating system businesses. These constraints include limits on certain contracting practices, mandated disclosure of certain software program interfaces and protocols, and rights for computer manufacturers to limit the visibility of certain Windows features in new PCs. We believe we are in full compliance with these rules. However, if we fail to comply with them, additional restrictions could be imposed on us that would adversely affect our business.
How it changed the behavior of Microsoft and others
So I would argue that even though the Department of Justice didn’t break up Microsoft, the lawsuit changed Microsoft’s behavior and made it a better corporate citizen. Apple needed time to figure out how to compete, and to compete on its own terms. Microsoft likely would have let Apple continue to exist, because Apple existing with a small part of the market made Microsoft’s legal life less complicated. The DOJ action gave Apple the time it needed to get a modern operating system onto the market, which then led directly to the products that transformed Apple.
Google, meanwhile needed a chance to establish itself. As long as Microsoft controlled the web browser, there was no technical reason Microsoft couldn’t make Google slower than Microsoft’s own search products. They could have easily inserted delays in the browser code that kicked in when you were using Google search. Post-2001, Microsoft didn’t dare do something like that, for fear of being found they had violated the Consent Decree.
Today, Google is a problem, and is facing its own antitrust actions. But I would argue that without the legal action against Microsoft, Google would have been more ruthless as a monopolist, if it had been permitted to survive in the first place. And an unrestrained Microsoft, handed a monopoly on search to go with its other monopolies, would not have been any better.

David Farquhar is a computer security professional, entrepreneur, and author. He has written professionally about computers since 1991, so he was writing about retro computers when they were still new. He has been working in IT professionally since 1994 and has specialized in vulnerability management since 2013. He holds Security+ and CISSP certifications. Today he blogs five times a week, mostly about retro computers and retro gaming covering the time period from 1975 to 2000.

Dave,
Another nice analysis. I’ll toss in a two other scenarios considering. The DOJ reaching a negotiated settlement agreement with MSFT after MSFT lost the initial ruling may not have been due to the change in political party in the White House. The reason is that reaching a negotiated settlement agreement is, historically, the most *likely* outcome after one party loses most or all claims at trial in a significant way and then appeals to an appellate court. Corporate lawyers like to say that trials involving a large business are really just extended (and expensive) negotiations – and this is especially true in anti-trust enforcement.
From opening an investigation, to issuing a preliminary finding to filing suit to contesting an appeal – every step in the process is anti-trust enforcers increasing their leverage in the negotiation with their target. The DOJ would *always* prefer to reach a negotiated settlement because it’s faster, cheaper, easier to enforce and far more certain. The time period during the appeal after the losing a significant trial judgement is usually the point of maximum leverage for the winning side. Historically, the appellate process usually reduces the penalties or remedies in some way. By definition the appellate process can’t make the penalty any *worse* for the losing side. The rare best case for the winning side is to just get what they already “won”, except three or four years later. It’s quite likely, if not probable, that even if the democrats had won the presidential election, the DOJ would have reached a similar negotiated settlement with MSFT during the appeal.
The second thing worth considering is an alternative hypothesis to explain why anti-trust enforcement against mega-corps (especially in high tech) often doesn’t end up mattering much. You mentioned the “Free Market Competition Will Fix It Anyway” hypothesis but, to me, an even more likely scenario is “Disruption Nerfs Mega-Corp Monopolists More than Competition and Anti-Trust Combined.” I’m referring here to Disruption Theory as outlined by Harvard economist Clayton Christensen in his “Innovator’s Dilemma” books. During the relevant time period (1998 – 2005), neither Google nor Apple seriously even *tried* to compete with MSFT in desktop operating systems or Office suites (MSFT’s strengths) and, more importantly, MSFT completely missed prioritizing Search and Mobile until Google and Apple were already dominant in each respectively and it was too late to mount a serious threat. Pocket PC, Zune and MSN Search were not only too late, they each managed to miss the point in their own uniquely dumb ways – a timeless trend which big mega-corps cluelessly trying to cover all possible bases manage to keep repeating.
Note: The Chrome Browser was only ever a long-term loss leader to help support Google’s real business: search engine advertising. Google would have been happy to pay MSFT billions a year for premium placement in the Internet Explorer browser just like they paid (and still pay) Apple billions for premium placement in Safari browsers. Pocket PC and Windows Mobile were each focused on fighting “the last war” instead of the next war (Pocket PC was clearly conceived with a PDA mindset more akin to Newton / Palm Pilot and Windows Mobile was likewise targeted at Blackberry, Nokia and Palm Treo phones. As Christensen predicts, neither iPod / iPhone nor Google Search ever *beat* MSFT in a strength-on-strength battle – they reshaped the battle field in ways which make the giant’s strengths that won the last war *irrelevant* in the next war. Ultimately, in high-tech, disruption only takes time (7-10 years) to weaken incumbents who grow too large and too successful on monopoly-scale franchises. Since anti-trust also takes 7-10 years to win against a mega-corp determined to drag out all appeals. Either way, the success rate in taking down monopolistic leaders is around 70%-80% but anti-trust is expensive and uncertain, whereas the inevitable beast of Disruption is inexorable, automatic and free.
Of course, I’m not arguing anti-trust enforcement isn’t needed or can’t do good things, just that specifically in high tech, it’s not the only Horseman of the Apocalypse all tech monopolists fear (nor the one they fear *most*).