Michael Dell Has Spent a Decade Buying One Thing Back: Control

In February 2013, the man whose name is on the building offered to buy it back. Michael Dell and the private-equity firm Silver Lake agreed to take Dell Inc. private at $13.65 a share, a deal valued at roughly $24.4 billion — the largest leveraged buyout since the 2008 crisis.¹ On paper it read like a tired PC maker fleeing a punishing public market. That was the cover story. What actually happened over the next decade is that the founder spent tens of billions — much of it borrowed — to do one thing, again and again, in different costumes: take control, then defend it.

The official narrative is that Dell is a hardware company that restructured to survive the cloud era. The truer reading is that Dell is a founder-control machine that happens to sell servers and laptops. Every capital transaction since 2013 — the LBO, the EMC mega-deal, the strange non-IPO of 2018, the VMware spin — is a move on the same board, and the prize on every move is the same: ownership concentration in one man's hands.

Going private was never about the PC business

Read the 2013 buyout as a turnaround and it looks reasonable but uninspired. Read it as a control play and it snaps into focus. A public company answerable to quarterly expectations cannot easily take on a transformation that gets worse before it gets better. A private company controlled by its founder can. The buyout removed the audience. It let Michael Dell stop performing for the market and start rebuilding the cap table on his own terms — which is precisely what he did next, at a scale no public board would have stomached without a fight.

The EMC debt was the point, not the price

In October 2015, barely two years private, Dell announced it would buy EMC for $67 billion — at the time the largest acquisition in the history of the technology sector.² The deal closed in September 2016 and created Dell Technologies on the back of roughly $45.9 billion of newly issued debt.² The financing tells you who was steering: it leaned on a syndicate of banks and a structure that minimised equity issuance — though the deal did include a tracking stock (DVMT) issued to EMC shareholders, a concession forced by the sheer size of the price tag.² Even so, the tracking stock carried no direct claim on Dell's voting equity. Dell had already demonstrated his preference for debt over dilution in 2013, when the LBO was financed partly by a $2 billion loan from Microsoft rather than new shares.¹⁰ The instinct was the same at EMC scale: protect the cap table, pay in dollars rather than votes, and let the balance sheet bear the cost.

This is the mechanism that runs through everything. Debt is expensive and dangerous, but it is silent — it doesn't get a board seat, it doesn't dilute the founder, it doesn't show up at the annual meeting. Equity is cheap when your stock is strong, but every share sold is a sliver of control handed to someone else. Dell, over and over, chose the path that protected the cap table even when it loaded the balance sheet. The structural choice was the strategy.

The 2018 "IPO" that wasn't

Dell's return to the public markets in December 2018 is widely misremembered as an IPO. It was nothing of the kind. The company completed a Class V transaction — paying $14 billion in cash and issuing about 149.4 million new Class C shares to buy out the DVMT tracking stock that had traded since the EMC close.³ Class C shares began trading on the NYSE that day.³ The distinction matters because an IPO is a company asking the market to set its price and, implicitly, to share control. This was the reverse: a controlled re-listing engineered to bring Dell back to public markets on Dell's terms, with the founder's grip intact. The label everyone reaches for — "Dell went public again" — is exactly the label that obscures the move.

The re-listing came at a real cost, and it was not borne by the founder. In November 2022, Dell agreed to pay $1 billion to settle a Delaware lawsuit brought by former Class V stockholders, who alleged the 2018 transaction shortchanged them by billions against fair value.⁷ That is the price of a control-first structure made visible: the minority holders along for the ride argued they paid for the founder's convenience. A billion dollars is what it took to make that argument go away.

Spinning off VMware to simplify what he already ran

On November 1, 2021, Dell completed the spin-off of its 81% stake in VMware as a special stock dividend, handing shareholders 0.440626 VMware shares for each Dell share they held.⁴ On its face this looks like the opposite of empire-building — giving away a crown jewel. But control logic explains it cleanly. The EMC debt needed servicing and the structure had grown baroque; spinning VMware lightened the load and simplified a corporate structure Michael Dell already dominated. You don't surrender control when you spin off a subsidiary you control through a parent you control. You tidy the board so the one seat that matters has a clearer view.

Isn't this just a founder protecting his baby?

The honest objection is that none of this is doctrine — it's sentiment. Plenty of founders cling to their companies out of ego, and you can fit any sequence of deals to a tidy thesis after the fact. Fair. But the control read survives a test sentiment fails: it predicts the financing choices, not just the emotion. A founder who merely loved his company would still raise equity when it was cheap and dilute himself to de-risk a $45.9 billion debt load. Dell didn't. And the structure has a brute-force dimension that the proxy's careful language can obscure: Class A shares (held by Michael Dell) and Class B shares (held by Silver Lake) each carry ten votes to every one vote for the publicly traded Class C — a super-voting architecture that, after the EMC close, gave Dell alone roughly 73% of the company's voting power.⁹ Today, with Silver Lake having converted and sold down much of its Class B stake, control rests on a combination of roughly 46% economic ownership⁵ and the residual super-voting weight of remaining Class A shares — a layered grip that is harder to dislodge than a simple head-count majority. That's a harder thing to engineer and an easier thing to defend in court, which is exactly why the company spends ink in its proxy justifying it.⁶ Sentiment doesn't file paperwork that careful. Doctrine does.

Strip away the deal jargon and the picture is almost monotonous in its consistency. Take the company private to escape the audience. Borrow rather than dilute to buy scale. Return to public markets through a side door that keeps the controls. Spin off a jewel to simplify what you already command. Through all of it, the founder's stake sits at roughly 46% with sole voting power — the number every move was quietly serving.⁵ Michael Dell didn't build a hardware empire and happen to keep control of it. He built control, and let the hardware empire be the thing it happens to own.