The 7% discount now has a ceiling

Two weeks ago Medtronic’s exchange offer for its MiniMed stake looked like a flat 7.53% discount with three weeks of runway. The terms have not changed by one word. The discount has: at Friday’s prices it is 3.55%, because a single number buried in the prospectus has switched on. That number is the upper limit, and every split-off exchange offer has one.

The lesson

How the upper limit in a split-off exchange offer works

A split-off is how a parent company hands out a subsidiary it has already floated. Instead of a cash tender, it offers its own shareholders shares in the subsidiary, at a discount, and lets them swap. The swap is not priced on the day you tender. It is priced on an averaging period near the end of the offer, from volume-weighted average prices, and it is then capped at a maximum number of subsidiary shares per parent share.

Those two features fight each other. The discount is expressed in money. The limit is expressed in shares. As long as the limit is slack, you get the money. The moment it binds, you get the shares — and the money shrinks.

The same document says how the two prices are counted: the exchange ratio “will value Medtronic Ordinary Shares at 100% of the Average MDT Price and will value MiniMed Common Stock at 93% of the Average MMED Price.” That 93% is the discount. Divide by it and you get the $107.53 per $100. So the ratio the formula asks for is the parent price divided by 93% of the subsidiary price — and then it is cut back to 4.5939 if that comes out higher.

Friday’s closes, run through it:

Two derived numbers fall out of the same formula, and they are the ones worth carrying. The limit switches off above a MiniMed price of $88.64 ÷ (0.93 × 4.5939) = $20.75; above that the full discount comes back. And the swap stops making money altogether below $88.64 ÷ 4.5939 = $19.30, which is 3.4% under Friday’s close. Between $19.30 and $20.75 you own a diluted discount that gets thinner as MiniMed falls.

This is where the mistake lives. The prospectus prints its own illustration: had the averaging period been 9 to 11 September, the ratio would have been 4.3599 — comfortably under the limit, full discount intact. MiniMed closed at $22.50 that day. It closed at $19.98 on Friday, down 11%, while Medtronic fell 3%. Nothing about the offer changed; the ratio the formula produces moved from 4.36 to 4.77 and crossed the limit on the way. Anyone who read the illustration as the deal, rather than as one dated example, is counting twice the return that is there.

The averaging period is the other half of it. The final ratio is set from the three trading days ending on the second trading day before expiry — here 5, 6 and 7 October — and announced by 9:00 a.m. New York on 8 October. Three days of the subsidiary’s price, chosen by the calendar, decide the whole trade, and there is no lower limit to catch you if it goes the other way.

This week’s lesson has its own guide: split-off exchange offers and the upper limit. Earlier guides: odd-lot priority (issue #1), the Canadian paid-up-capital test (issue #2) and broker cutoffs and record dates (issue #3).

This week’s situations

Live offers with terms and a deadline, each worked through with the same arithmetic. No new situation came out of this week’s filings: thirty-odd issuer tender offers were filed and every one of them was for shares in a fund that has no listing to buy. These carry over, re-priced on Friday’s closes.

Two more offers are open without a usable window. Fnac Darty (Paris) trades at €34.50 against a cash offer now worth €35.00 after the €1 dividend paid in June, +1.45%, but the closing date is set only after European merger clearance, notified on 14 September (opening release). Eurobio Scientific (Paris) trades at €25.10 against an announced buy-out offer at €25.30 to be followed by a compulsory acquisition, +0.80%, with no calendar filed yet (intention release, 16 September). An offer without a dated holder window is a price, not a deadline, and it cannot be worked. The author holds none of these.

Gross = (offer − price) ÷ price, on the part of the position the offer takes up. Annualised = gross ÷ days to expected cash × 365, for that trade only — not a yearly or portfolio return.

Broker execution

Four measured gaps, and three deadlines nobody has published

This is the running record of what a broker actually does, as opposed to what the offer document says. Arbutus: four hours early. Oma Savings Bank: about two days early. Delivery Hero: two days early. Medtronic: the depositary’s book-entry channel shuts seven hours before the offer’s midnight expiry, and the broker’s own cutoff sits somewhere before that — asked for, still not answered, three weeks into a four-week offer.

For Medtronic, ACV Auctions and Nagarro the cutoff is simply unpublished. There is no pattern in the four measured gaps to extrapolate from — four hours and two days come from the same broker on offers of the same kind — so the only sound method is to ask on the day an offer opens and to treat the depositary’s deadline as the outer bound, never the offer’s.

One practical note from this week’s account: a limit order for a London-listed line is entered in pence while the position feed reports pounds. Price from a live quote in the units the quote comes back in, not from the feed.

The dates that matter · New York time

Killed, and what to look for next time

A perfect offer with a 24% hole in it

Utah Medical Products filed the cleanest small self-tender of the month on 24 September: a fixed $75.00 in cash for 650,000 of its 3,173,818 shares, expiring 17:00 New York on Tue 7 Oct, with odd-lot priority spelled out — “Odd Lots will be accepted for payment before proration” for a holder of fewer than 100 shares who tenders all of them. The shares closed at $74.18 on Friday, +1.11%. Then the tax section: “Even if a foreign stockholder has provided the required certification to avoid backup withholding, the Depositary will withhold United States federal income taxes equal to 24% of the gross payments payable to a foreign stockholder.” On a 99-share tender that is 24% of $7,425, or $1,782 withheld, against a spread of $81. It is recoverable only by meeting one of the tests in the tax code and filing a United States return. Next time: read the “foreign stockholder” paragraph before the discount, and read it on operating companies as well as on funds — this is the American twin of the paid-up-capital sentence in a Canadian issuer bid. Offer to Purchase, 24 September 2026. Author holds: no.

Two mechanical kills worth naming. Clywedog Therapeutics has run a fixed-price tender with real odd-lot priority since 15 September and there is still no tradable line for it at the broker under any of its names: an offer you cannot buy into is not a situation. And Computer Modelling Group, whose substantial issuer bid expired on 21 September, has now appeared four times in the filings feed on the strength of results and completion headlines. A results release is not a new offer; the date on a feed is the date it was captured, not the date the thing happened.

The bigger kill is a category. More than thirty issuer tender offers were filed this week and nearly all of them came from interval funds, non-traded business development companies and private credit vehicles repurchasing their own units at net asset value. The terms often look generous. There is no listed share to buy, so there is nothing to do.

And the one from issue #2 finished

Air Canada’s C$800 million substantial issuer bid closed on 25 September. It cleared at C$29.00, the floor of the C$29.00–33.00 range; 66.8 million shares were deposited against 27,586,206 taken up, so holders who tendered at or below the purchase price were prorated to about 41% — “other than ‘odd lot’ tenders, which are not subject to proration”. That is the odd-lot mechanic doing exactly what issue #1 described: 99 shares would have been taken in full while a thousand-share holder received cash for 410 of them. It is also the paid-up-capital lesson doing exactly what issue #2 described. Paid-up capital was about C$10.60, so at a C$29.00 clearing price the deemed dividend for a non-resident was C$18.40 a share, C$2.76 withheld at the treaty rate, against a spread of C$0.97 over the entry price. The position was sold in the market on 11 September at C$27.80 for a realised loss of C$24.95 rather than tendered. Preliminary results, 25 September 2026. Author holds: no.

The book

Every position, and every closed one

Open, at Friday’s closes: Arbutus, 99 shares at an average cost of $5.0501, tendered at the purchase price, awaiting the auction result. Apollo Commercial, 300 at $6.9150, held through the vote and the first distribution. Oma Savings Bank, 300 at €17.07, all 300 submitted into the €17.20 cash offer, which closed on 25 September; the final result is expected around 30 September and settlement within five banking days after it. Delivery Hero, 50 at €37.035, not yet accepted. Perfect Corp, 1,200 at an average cost of $1.8088, held into the October meeting. And 100 contingent value rights from the Atai transaction, which still sit on a placeholder line with no market.

Closed since the last issue: Lindsell Train Investment Trust, 260 shares sold on 22 September at 588p for a realised gain of £22.67 on a £1,492 position. The tender it was bought for turned out to be open only to holders on the register at 18:00 London on 14 September, and London settles two business days after the trade, so shares bought on the day of the record date were never eligible. That correction was printed in issue #3; this is how it ended.

Realised losses still on the record: Air Canada, −C$24.95, and Zevia, −US$323.10.

The outcome record, every closed position with fills and realised P&L: https://99shares.org/scoreboard/

This publication discusses the mechanics, deadlines and public filings of corporate actions such as tender offers, exchange offers and liquidations. It is not a financial advisory service under the Financial Advisers Act 2001. The author is not a licensed financial adviser and is not an appointed representative of one. Nothing here is a recommendation to buy, sell or hold any security, and nothing is tailored to any reader’s circumstances. We do not answer individual investment questions. We accept no payment, sponsorship or other benefit from any issuer named. Any position in a named issuer is disclosed in the issue. Figures are taken from the primary source cited. Readers should not rely on this publication in making an investment decision.

Terms of use · 99shares.org · hello@99shares.org · Sent every Sunday. Unsubscribe at any time.