The way a stock buyback works is:

1. You have a company worth \$1,100, with \$100 of extra cash on hand.
2. It has 110 shares outstanding.
3. Each share is worth \$10.
4. You decide you don’t need all that cash and that you should give it back to shareholders who want to sell.
5. You buy 10 shares for \$10 each, using the \$100 in extra cash.
6. Now you have 100 shares outstanding, no extra cash, and a company that is worth \$X.

What is X? In this schematic description it seems like it should be \$1,000. You had a company worth \$1,100. It gave \$100 back to shareholders, in exchange for nothing. (In exchange for their shares, yes, sure, but as far as the company is concerned those aren’t a thing.) \$1,100 minus \$100 is \$1,000. There would be something odd about getting any different answer. Before the buyback the shareholders, collectively, owned \$1,100 worth of stuff. (Their stock.) After the buyback the shareholders, collectively, owned \$X + \$100 worth of stuff. (The cash, plus the remaining stock.) If \$X is more than \$1,000, then the buyback—just shuffling around ownership of the \$100—created value; if it’s less than \$1,000, then it destroyed value. In an efficient world you shouldn’t be able to create value out of thin air just by changing who holds on to some cash.

If that’s true, then the buyback won’t reduce the value of the company by the amount of the buyback. It will reduce it by some lesser amount. It might even—and it is hard to tell, hard to disentangle the effect of the buyback from the effects of announcements of good news, etc.—but it might even leave the value of the company unchanged. In my example, the company would be worth \$1,100 after the buyback, and the shareholders would get a “free” \$100. The implication there would be that the \$100, when it sat on the company’s balance sheet, was worth \$0: The company had no productive uses for it, and the market expected it to just waste it on garbage, so returning it to shareholders created \$100 of pure windfall value.

I suppose there is no reason that the buyback couldn’t even increase the value of the company—it could be worth \$1,200 after the buyback, even after taking out the \$100 in cash—but that would be a little silly. The implication there would be that the cash on the company’s balance sheet was worth less than zero, that the market expected the company not just to utterly waste the \$100 but to waste it in a way that destroyed additional value. I suppose in small quantities and at some margins this could happen, but it can’t really scale. If you have a \$100 billion company and spend \$10 billion buying back stock and that—that alone—makes the company worth \$110 billion, then you’ve created a perpetual-motion machine. You can borrow money to buy back all the stock except for one share, but that share would be worth \$200 billion. (Then you sell a 50% stake in that share, use it to repay your loan, and bang, free money. I realize the math here is a bit imprecise.)

Anyway here is a claim that Apple Inc. reached its \$1 trillion valuation by buying back stock:

Apple’s recent success on Wall Street isn’t due to its technological innovations or its sleek products. Instead, its stock has been juiced by a record-breaking number of buybacks, in which the company buys shares of its own stock, causing the supply to drop and the price to rise. In May, several months after Congress passed a massive corporate tax cut, Apple pledged \$100 billion to stock buybacks in 2018—and is halfway to that goal. With \$285 billion in cash on hand, it can afford to buy even more.

Viewed over a period of decades, a number of products and achievements played a role in getting Apple to where it is today. But as the company’s profit margins have shrunk, stock buybacks played a crucial role in getting Apple over the trillion-dollar finish line first. This asterisk should be something of a scandal. Apple is the poster child of the current spate of stock buybacks, which are starving investment and exacerbating inequality.

I feel like … nope? As a matter of the “poster child” thing; Apple spent \$11.6 billion on research and development in fiscal 2017, up from \$10 billion in 2016 and \$8.1 billion in 2015—a bigger two-year increase in R&D dollars than most other companies spend on R&D total. You should entertain the possibility that Apple’s vast research budget pays for the research that Apple wants to do, and that some of the unusable excess gets returned to shareholders in the form of buybacks.

But mostly nope as a matter of arithmetic. The theory here is essentially that if Apple had 220 billion more dollars—roughly the amount of buybacks it has done in the last six years—then it would be worth less money. It doesn’t sound especially plausible. Presumably Apple plus \$220 billion would be worth at least, like, a dollar more than Apple alone? And if the buybacks work so well, why not do more? Why not sell off all of Apple’s businesses, use the money to buy back stock, and leave Apple as an empty shell with a reduced share count and a \$2 trillion valuation?

There is a widespread view among critics, not only of buybacks but of financial capitalism generally, that it is all tricks and that the tricks are easy. This article is titled “Apple’s Stock Market Scam.” The idea is that investors are incredibly easily deceived, that stock prices reflect no reasoned judgments about a company’s business prospects, that gaming those prices is child’s play. It is a world in which investors are entirely unable to evaluate a company’s business, and can be tricked by devices—stock buybacks, non-GAAP accounting—that occur in broad daylight and are subject to frequent intense criticism. It would probably be very convenient for CEOs if they could increase their valuations that easily, but I suspect it’s not true.

Who controls a company?

I don’t know anything really about Chippewa Capital Partners LLC or Mesabi Metallics Co., but just let this prose wash over you:

This action is necessary because defendant Clarke has gone rogue. Without any authority, Clarke has purported to remove the other two directors in each of Chippewa and Mesabi, and anointed himself as the sole director, chief executive officer, spokesperson and leader of those companies, in which he has no ownership or contractual rights. Furthermore, he proceeded to announce these unauthorized actions to the press, employees of the two companies, and key third parties such that those parties believe that he alone has authority to take the actions on behalf of the companies. The sole owner of Chippewa and Mesabi is Nubai and it removed Clarke from all positions and offices held by him in both companies pursuant to the operating agreements of those companies. However, Clarke has simply ignored the written resolutions removing him and instead arrogated to himself all the authority of sole owner, sole director, sole officer and sole spokesperson.

That’s from a declaration by Nubai Global Investment Limited, a company that claims to own Chippewa and Mesabi, which run an iron-ore mining project in Minnesota, in its lawsuit against Tom Clarke. Tom Clarke is … was? … the chief executive officer of Mesabi and Chippewa. Nubai tried to remove him. He wasn’t having it, so he fired Nubai’s two directors instead. Here’s his view of the matter:

"They were supposed to be \$250 million dollars into Mesabi by April 15th. They did not do that. I have been pushing them every single day. It's been more than three months now. So we terminated or removed their two directors on Sunday." Clarke said if they don't pay the \$90.5 million that is still owed, he'll bring new partners on board, from North America.

He maintains he is now the sole board member of Chippewa Capital. "We have very solid legal advice. We have all of our creditors behind us. We've communicated with all key stakeholders. Everybody knows what's happening," he said.

Here’s Nubai’s view of that legal advice:

Susan Fennessey, Mesabi’s General Counsel (“Fennessey”) advised Clarke on July 25, 2018 that the resolutions of Chippewa’s and Mesabi’s sole owners were valid and enforceable. In response, Clark immediately cut off her access to the Mesabi computer systems and email server and thereafter sent her a notice suspending her from her role as General Counsel (the “GC Suspension Notice”). Nubai has informed Fennessey that she is not suspended and has requested her to assist with various essential tasks in the last week. Fennessey, however, has been unable to act because she has been shut out by Clarke from the company’s computer systems and emails.

And its view of those communications with stakeholders:

Clarke had meetings scheduled for July 26, 2018 with Minnesota Governor Mark Dayton and Itasca County Attorney Jack Muhar, related to the Project. In the GC Suspension Notice, Clarke forbid Mesabi’s General Counsel from attending those meetings.

On July 26, 2018, despite having been removed from his positions in Mesabi by the sole owner, Clarke went ahead and met with Governor Dayton and County Attorney Muhar. Upon information and belief, Clarke sought to mislead both by representing to them that he maintained an equity stake in Chippewa and was the sole director and CEO of Chippewa and Mesabi.

It is conventional to think that the ownership and control of a company are essentially black-and-white legal matters that are set forth clearly in its governing documents. (Certainly Nubai’s declaration makes it sound like the ownership of Chippewa and Mesabi is fairly clear, though to be fair we don’t have Clarke’s side of it.) And in the rare cases of ambiguity, you go to court and the court clarifies who owns and controls the company. (Certainly that is Nubai’s approach.)

But as we like to talk about around here, these are essentially practical questions and there are other, faster ways to answer them than going to federal court. For instance, if you meet with the governor on behalf of the company, and say you run the company, and he believes you, then he is not going to be well pleased if half an hour later someone else traipses in and says “no actually run the company.” Like, being the first CEO to get in front of the governor gives you a lot of points in the battle to actually be the CEO, at least in the governor’s eyes.

Or similarly if you control all the computer passwords, and the board fires you, and you ask the general counsel “can they do that,” and she says “yep they just did”—well, you can just change the password so she can’t get into the computers, and you have solved your general-counsel problem. Duly appointed officers without access to the computers have less practical authority than improperly appointed officers with the right passwords. Nubai’s complaint demands that Clarke “disclose all usernames and passwords held by him relating to Mesabi or Chippewa,” because just getting the court to declare him Not CEO doesn’t have that much practical significance. The legal formalities only get you so far; what you really need is to get the keys back from him.

Goldman vs. gold.

Oh no:

Mr. Thornton, 64 years old, spends only two or three days a month at Barrick’s Toronto headquarters. He communicates constantly with his team by email, sometimes firing off requests late at night or when executives are on vacation, according to people familiar with the matter. If he doesn’t receive a quick response, Mr. Thornton repeats the request with the new subject line: “resending,” these people said.

That’s from this Wall Street Journal story about how John Thornton, formerly a senior executive at Goldman Sachs Group Inc. and now the chief executive officer of Barrick Gold Corp., is running that company more like an investment banker and less like a gold miner. Some of that is about financial discipline, but some of it is cultural, and I feel like there are a lot of industries where you’d be sad if your new boss was a former investment banker. I am sure that this “resending” thing is not unique to investment banking, and frankly it’s a bit extreme even there, but, man, I don’t know. In investment banking the job is to sit around responding to emails at all hours. In gold mining, I gather, part of the job is to go deep beneath the earth and scrape metal out of rocks. Like, sometimes people are busy, you know? If I got one of those “resending” emails I would be very tempted to reply “sorry I didn’t reply earlier I WAS DEEP DOWN IN A GOLD MINE.”

Obviously I realize that the executives probably weren’t really in gold mines when he emailed, but I don’t know, it does seem like a less sedentary and email-focused job than banking.

Anyway on the broader substantive points I am … not sure how to take Thornton’s results? Like basically under Thornton Barrick has become a smaller gold producer—with fewer mines and 25 percent lower production—but arguably a better company:

After pushing Barrick to cut costs and get rid of less productive mines, Mr. Thornton has more than halved Barrick’s approximately \$13 billion in debt. The company has become more profitable, reporting an adjusted profit of \$876 million last year, compared with a loss of more than \$10 billion in 2013.

You could imagine that being a good tradeoff, especially from his perspective: He’s a banker running the company like an activist shareholder; he cares about cash flows, not gold production. “The most successful companies in history all emphasized free cash flow as their single most important metric,” he told the Journal. “We do not accept that gold mining deserves some special kind of exemption.” The upside of hiring a banker to run your gold company is that he is not beholden to the prejudices of gold miners. The downside is that it is always possible that different industries are different from each other, and do deserve some exemptions, and that financial optimization really is more central in the financial industry than in gold mining. In any case the stock is down about 37 percent since he took over.

Things happen.

Dell Projects Revenue Boost in Pitch for Tracking-Stock Buyout. Alcoa Requests Reprieve From Trump’s Aluminum Tariffs. Cigna ‘Dramatically Overpaying’ for Express Scripts, Icahn Says. Facebook to Banks: Give Us Your Data, We’ll Give You Our Users. Turkey’s Cratering Markets Stoke Speculation Over Extreme Measures. “Only about 2% of the people with devices that use Amazon’s Alexa intelligent assistant—mostly Amazon’s own Echo line of speakers—have made a purchase with their voices so far in 2018.” Drug Deals and Breakups: Venmo Social Feed Rankles Privacy Advocates. SoftBank Weighs the Largest Public Listing Ever. Banco Popular: new report says alternatives to rescue were worse. Banks After Tax Cuts: Loan Growth Slows and 3,200 Jobs Disappear. Gary Cohn Is Happy in the Hamptons, Golfing and Crypto Investing. “Sue Gross forked over \$37.8 million for pricey pads in Laguna Beach’s exclusive Irvine Cove community, where Bill spent most of his adult life — just so he couldn’t get his hands on them to begin putting down new roots there, according to court papers and sources.” Move Over Millennials, It’s Gen Z’s Turn to Kill Industries. “A small degree of substitutability between television viewing and sexual activity.”

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Matt Levine is a Bloomberg Opinion columnist covering finance. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz, and a clerk for the U.S. Court of Appeals for the 3rd Circuit.