JF Tardif — "We're Definitely Ready" If Markets Turn Ugly: cheap growers on the long side, an options overlay on the short side

This is the single most useful idea in the interview, and it is about portfolio construction rather than any stock. A put option is a contract that pays out if an index falls below an agreed level before an agreed date; you pay a small premium for it whether or not it pays out, the way you pay an insurance premium. Tardif's fund owns more shares than it has sold short — it is "net long" — and then buys enough S&P 500 puts on top that, if markets fall, the whole book is effectively positioned below zero net exposure.

The reason he switched to this in 2021 is that shorting individual companies became painful and unpredictable: a short position's losses are unlimited if the stock keeps rising, and the last few years punished anyone doing it. A put has the opposite shape. The most you can lose is the premium — "a very small cost every month" — while the payoff grows non-linearly the faster and further the market falls. That asymmetry is why he can afford to stay hedged permanently rather than trying to time the top.

The discipline around it is what makes it work: he buys more puts after markets rally (when protection is cheap and least wanted) and sells them into declines (when it is expensive and everyone wants it). That is the reverse of what fear and greed prompt, and it also produces cash exactly when he wants to buy stocks. Asked whether this is a rainy-day position, his answer is the episode's title: "we're definitely ready if something really bad happens quick. We're definitely protected."

SOXX is a fund that owns the big semiconductor companies as a basket. Tardif is short it — he has sold borrowed units expecting to buy them back cheaper — and he explains why he uses the basket rather than naming individual chip stocks: he does not want to disclose single-name shorts, and an index removes the risk that one company surprises him with good news while the sector rolls over.

The argument is arithmetic, not sentiment. Nvidia has guided to 70% growth next year. If the whole AI supply chain grows at that rate, total spending on this build-out reaches roughly $1.7 trillion a year. He simply does not believe an economy the size of the world's can sustain spending at that level indefinitely — so somewhere ahead is a peak, and after a peak, spending does not plateau, it declines. Chip company earnings today are the mirror image of that spending, so when it turns, so do they.

He is careful about two things. First, timing: "it could be in 27, maybe it's in 28. That's a real debate. In my mind it's not if." Second, sizing: he adds to the short when the index rallies and covers half of it after a 10-15% fall, so the position is managed against the tape rather than held to a target price. He does not need a crash — a correction is a perfectly acceptable outcome.

The Canadian banks have just reported excellent results, and Tardif's point is that this is precisely the problem. Very little of the recent strength came from ordinary lending. It came from trading desks, wealth management, portfolio management and insurance — businesses whose profits rise and fall with markets. Paying a record multiple for peak earnings from cyclical businesses is the classic way to lose money slowly in a "safe" stock.

He takes on directly the popular argument that the banks have become structurally better — more diversified, more fee income, less loan-dependent. He grants the diversification, then points out that the fee income is not the stable kind people assume: in a bear market, assets under management shrink, and in a 20% drawdown trading volumes fall too. So the supposedly defensive earnings decline at exactly the same moment the loan book deteriorates.

Then he adds the risk nobody is being paid for: political noise about Canadian banks' access to the US market. He does not expect anything to come of it, but "you're not paying a discount for that risk. Right now people are paying the highest premium they have ever paid." His practical advice, given to a friend's father, is unusually concrete: if the shares sit in a tax-free account, selling costs nothing in tax, so harvest the gain and buy them back later — because six months or a year from now they will likely be "at the same price or lower." He confirms Timelo is permitted to short banks and does so, while declining to say which.

The popular argument runs: AI needs data centres, data centres need electricity and wiring, wiring needs copper — so buy copper miners as a safer way to own the AI theme. Tardif's answer is that the market has already made that trade, which means copper stocks are now an AI stock with an AI valuation and an AI downside. He points at a single day's tape as proof: three quarters of the market fell, and copper miners rose 10%.

His objection is a valuation comparison rather than a view on the copper price. On every measure he uses — price relative to the value of the assets in the ground, price relative to cash flow, price relative to earnings — copper equities trade far above gold equities. He owns a little copper and shorts some, which is a relative bet: if the whole complex falls, the expensive half falls further.

And the trigger is the same one behind his semiconductor short. Copper demand from the AI build-out is capital spending, and capital spending is exactly what he expects to peak and then decline. "Copper stocks are likely to go down once we go down the other side of the mountain and capex starts to go down. And it could be in 27, maybe it's in 28. That's a real debate. In my mind it's not if."

The other half of the pair. Gold miners, on the same valuation measures where copper miners look expensive, are "extremely cheap." That is the whole relative case — he is not forecasting a gold price, he is buying the cheaper of two mining sectors and shorting the dearer one. Most of his fund's net long exposure, he says, comes from gold and oil.

How he owns gold is as important as that he owns it. Rather than the large producers, he buys small companies with a queue of identifiable events ahead of them: a permit due, a preliminary economic assessment due, a final investment decision due. Each of those events, if it goes well, re-rates a small company much more than a good quarter re-rates a large one — and crucially, he can list them in advance, so he knows what he is waiting for.

He also treats gold exposure as something to hedge rather than sell. After a big rally he shorts more gold names; after a pullback he covers. The purpose is to avoid ending up with too much of a good thing after it has already run — "not to have too much gold especially after a big move up" — rather than to bet against the metal. The gold names he shorts are described by type, never named: "larger companies, no growth, maybe some of them not well managed."

His single highest-conviction gold position, and a recent purchase, is a small Ontario gold company worth under $300 million. The interesting part is not the company but how he got comfortable with it, because mining juniors are the easiest place in markets to lose money on a good story.

Mining projects fail for technical reasons that are genuinely hard for an outside investor to assess: the geology can be misread, the rock may not give up its gold economically (that is what "metallurgy" means — whether the processing actually works), or the engineering may not stand up. Tardif's shortcut is to follow a handful of specialist investors who have already done that work and put their own money in — he names Michael Gentile, "Cohen at Scotia," and Lassonde. If they have vetted the geology, the ore and the metallurgy, "these projects are probably going to be mined," and he inherits their diligence.

The second half of his answer is why he prefers this to simply buying a large, safe producer like Agnico Eagle. In a small company he can enumerate the catalysts coming and he knows, or knows someone who knows, the board and management. That combination — a queue of known events plus a short line to the people responsible for them — is his stated edge, and it does not exist in a large-cap. Note the risk this carries: a sub-$300m junior can fall a very long way if a single catalyst disappoints.

Last year's pick and the episode's home run: $11 to about $35. What makes it worth studying is that the original call was a written-down calculation, not a hunch. The company earned $2 a share; he judged earnings would double to $4 over three to five years; he applied a 10x multiple; that gave a $40 target. The company got to $3 in one year, and the stock did most of the work already.

The reason to still hold it is that the same calculation now has more in it. Compression — renting out the equipment that pressurises natural gas so it can move through pipelines — is mid-expansion in the US, with capacity nearly doubling. In Australia the company is now the largest player, serving liquefied natural gas exports to Asia, which he describes as very stable demand. In Canada, new pipelines and LNG Canada's second phase are coming. So $4 of earnings may arrive sooner than three years.

Two further sources of return sit on top. First, a re-rating: energy services have been unloved for years, and "these type of stocks could trade at 13, 14 times," which on $4 is a $50-plus stock. Second, the company holds net cash and the founder-CEO has a history of making significant acquisitions and, importantly, of not making them when the price is wrong. Tardif is candid that if oil falls the whole sector falls with it — but argues that compression, Australian LNG work and Canadian pipeline construction carry on regardless.

Last year's second pick, up 54% and still held. The name gets mislabelled as a staffing company, which is why it was cheap. What it actually does is supply services and training to governments — including military training — alongside its staffing operations.

That distinction is the whole thesis, because Canada and Europe have both committed to spending materially more on defence, and a large part of that spending is not equipment but the training and services required to use it. Calian sits directly in that flow.

The valuation has already moved from about 10 times earnings to about 14, so the easy re-rating is behind it, but he argues earnings growth now takes over and continues "for actually years to come." Management has also publicly said it is looking to make more acquisitions, which is the next identifiable catalyst — the same pattern he looks for everywhere else.

ADENTRA distributes wood and building products — the materials that go into finishing houses and offices. The share price has gone nowhere for five years, which is exactly the kind of setup he says defines his style: "if the business performs and continues to grow and produce free cash flow, eventually the stock should rise."

The history explains the dead price. The company bought businesses during and just after COVID, which pushed its borrowings up; then US housing weakened, so the acquired businesses did not perform as hoped; so no further deals were possible while debt was paid down. That process is now finished — "the balance sheet has improved. Now they're ready to make the next acquisitions." The market has not repriced for that because nothing has happened yet.

Meanwhile you are paid to wait: 8.5 times earnings, six times EBITDA (a rough measure of operating cash generation before financing and tax), free cash flow every quarter, in a fragmented industry where being the consolidator is a proven strategy. The important discipline is what he refuses to underwrite: he is not assuming a housing recovery. If housing does turn, profit margins expand and that is a bonus he has not paid for.

Colliers is a commercial real-estate services firm: leasing and sales brokerage, property management, investment management and an engineering consultancy. The stock has fallen 42% in a year — while profits and analysts' forecasts have gone up. That gap is the entire opportunity, and its cause is a story rather than a number: investors fear artificial intelligence will make services businesses like this obsolete.

Tardif is openly sceptical of that fear and gives a historical rebuttal that transfers well beyond this stock. Two hundred years ago bridges were engineered on paper. Computers arrived, then computer-aided design software arrived, and each time one could have argued engineering firms were finished. Instead the volume of engineering work grew and grew, because better tools made bigger and more ambitious projects possible. He expects AI to do the same. On brokerage the point is even simpler: the internet has existed for decades and people still use brokers to find a lease.

What is left is 12 times earnings — near the cheapest the company has ever been — for a business still growing, still acquiring, still run by its founder Jay Hennick, and generating roughly an 8% free-cash-flow yield that gets reinvested into more acquisitions. His two-branch summary is the reason the risk/reward appeals: if commercial property stays flat you collect that 8% with growth; if the return-to-office recovery continues, "the earnings are going to go up a lot" and the multiple can expand on top.

A drive-thru coffee and energy-drink chain, and by his own admission not the kind of stock people expect from him. The rule behind it is one he states plainly: "I love buying companies that are growing at a fire sale."

The growth has two independent engines that add together. The company opens about 17% more stores each year, and the stores it already owns sell 5-8% more than they did last year — which puts total growth north of 20%. Food has only just been introduced and is not yet in every location, so there is another leg still to come, and the store base is small enough that the chain can plausibly double, triple or quadruple over many years.

The price is 42 times earnings, which he concedes is expensive — "it's just a restaurant." The defence is comparative: Starbucks trades at 34 times and is not growing at all; Cava, a fast-casual chain, trades at 92 times; Shake Shack at 52. And the arithmetic of growth does the rest. If earnings compound at 20%, they double in about three and a half years — so today's 42 times becomes an effective 21 times on the earnings the company should be producing by then, without the share price moving at all.

H&R is a Canadian property trust being broken up: GO Residential buys the apartment portfolio, Blackstone buys the industrial portfolio, and a company controlled by the founder's family takes some of the leftovers. Tardif calls it not a takeover but a "take under" — a deal done below what he thinks the assets are worth. The stock has drifted from around $11 at announcement to about $9.75.

He is unusually direct about the governance objection. Management is buying a meaningful slice of the assets from the company it runs, at a price "some people looks like cheap" — a conflict of interest whether or not anything improper occurred — and there may be tax complications for unitholders. He plans to vote against, though he is realistic that heavy retail ownership may push it through anyway.

The part worth learning from is that his vote and his trade point in opposite directions, deliberately. He owned few units when the deal was announced, watched the price fall, and started buying. If the deal completes at today's price you make money on the spread and you receive units in GO Residential, which he separately thinks is cheap — so you acquire a cheap asset at a discount to an already discounted price. If the deal fails, he believes the stock recovers on its own merits. "Either way, I'm comfortable either scenario in terms of the current price." He also notes what he thinks should have happened instead: H&R once spun out its mall business as Primaris with a new young CEO and it worked well, so spinning out divisions was an alternative to selling cheap.

GO Residential is the New-York-focused apartment trust buying H&R's residential portfolio, listed in Toronto in US dollars. Tardif's view of it is short and favourable: "GO itself looks pretty cheap here… they have issues but" — the sentence trails off, and no specific concern is named.

What matters is the route in. Buying H&R units below the deal's stated value means that, when the transaction closes and H&R holders receive GO units, you have effectively bought those GO units for less than their market price. "GO is cheap and then you get it under, even cheaper going through." That double discount is why he says "I don't think there's a lot of risk here."

The host discloses she owns GO directly, and notes at the top of the episode that both CAE and GO Residential are her own holdings.

CAE builds and operates flight simulators and trains pilots. Tardif calls it "almost a monopoly," which is the reason to own it at all: airlines have very few alternatives, and pilot training is not optional — it is legally mandated and repeated throughout a career.

The stock has done poorly and there is new leadership. He met both the new CEO and CFO and was "extremely impressed" — the CEO came from large US companies and is working through every part of the business methodically, "every week looking at different things." That is a positive, but it also sets the timetable honestly: fixing a business one piece at a time takes years, and he expects earnings to be roughly flat rather than growing for the next two.

Hence the sizing, which is the genuinely instructive part. He owns it, but deliberately small — "it's not a big position yet because I'm not sure it's ready to take off soon." Conviction in a business and conviction in its timing are separate decisions, and he sizes to the weaker of the two. On the fear that airline cost pressure limits the turnaround, he is dismissive: pilot numbers grow every year worldwide and training is compulsory. The defence side, though small today, benefits from European and Canadian military spending.

Cannabis has been, as the host puts it, a heartbreaker of a sector. Tardif's route into it is the least exciting possible one: the company that already makes money. Village Farms has net cash rather than debt, is profitable, and should earn 25-30 US cents next year against a share price around $3 — roughly ten times earnings, which is a normal price for a normal business, not a speculation.

The operating case is cost and geography. It has just completed a production expansion in British Columbia and is the lowest-cost producer in Canada — in a commodity business, being lowest-cost is the durable advantage. That extra output lets it win back Canadian market share, and it also sells into Germany, where the medical cannabis market is expanding rapidly and, importantly, is supplied by imports, which makes Canadian growers the natural beneficiaries.

Then two free options. Democrats are again discussing US federal rescheduling, which he does not count on. And because the company holds US, Canadian and European assets in one structure, he thinks it is "probably the best candidate to be taken over" — with Curaleaf's announced move on Aurora as evidence the sector is consolidating again. His upside is a multiple, not a story: if sentiment improves at all, "it could easily be at 15 to 20 times earnings… so it could be a double."

This position is not an opinion about cannabis or about Trulieve. He owns it because the stock is expected to be added to an index, and index funds must then buy it mechanically regardless of price — a predictable wave of forced buying he is positioning ahead of. "We own it because it's going to go in the index. So it's more of an arb."

It is worth separating this cleanly from Village Farms in any read-across. One is a valuation and operating thesis with a multi-year horizon; this is a technical trade with a defined event and, presumably, a defined exit. He offers no view on the company's earnings, balance sheet or prospects.

Magna mines copper and nickel in the Sudbury region of Ontario. It is not yet a position — he had only spoken to the company the previous week — but he describes it as promising, and the way he did the work is the lesson.

He did not read a broker report. He remembered someone from his Bay Street days, discovered the man was now Magna's VP of development, called him, and talked for half an hour. That is a repeatable research method for anyone with a network in an industry: find the person who is actually building the thing, not the person selling the stock.

What he liked matches his standing screen. The company produces from one area today and has a second, third, fourth and fifth to bring on — a queue of identifiable events, "lots of catalyst." And two sets of sophisticated money are already in: $100 million from an investor group that owns and operates mines itself (so it can judge the asset technically, and has cash flow to deploy), plus the Dundee group, long-standing mining investors. Both are the same "let other experts do the geological diligence" principle he applied to McFarlane Lake.

Toromont is the Caterpillar dealer for much of eastern Canada, plus a second division that has become popular because it sells into AI-related construction. The stock is up 40% in a year and the question put to him was hold or sell.

His answer separates business quality from share price completely, which is a useful habit. The company is excellent — "extremely well managed," a great balance sheet with roughly no net debt or possibly net cash, and management has been talking about acquisitions for a while, so a deal could arrive any morning. He simply will not pay the current multiple: "I just find it's too expensive. So me personally, we don't own it." After a subsequent pullback, still no: "still too expensive for me."

The volunteered detail is what makes this Neutral rather than Negative, and it is a discipline worth copying: "I'll reveal that we're not short it either." Expensive is not the same as shortable. A quality company with excited buyers and an acquisition pipeline is a dangerous thing to bet against, however rich the price.

Champion Iron mines iron ore in Quebec. Tardif looked at it precisely because it and Labrador Iron Ore Royalty are both, in the host's phrase he happily adopts, "in the toilet" — cheapness is what gets a name onto his desk. Then three separate checks failed.

First, supply: a major new mine is coming into production in Africa, adding roughly 5% to world iron ore supply. In a commodity, a 5% supply increase is a large number and it lands regardless of what any individual company does. Second, demand: he is "not super keen" on the economy, and iron ore feeds steel, which is as cyclical as industry gets.

Third — and this is the one that decided it — the founder has been selling shares. He is careful to be fair about the man, who built a well-established company and made himself a lot of money doing it. But the signal stands on its own, and it is the same screen that made him pass on Propel: the people with the best information are reducing. So he is "staying away, but I'm watching" — waiting to see iron ore bottom and turn before revisiting. On tariffs he notes they are not the issue here, since both companies sell mostly into Asia and Europe.

If he were to own iron ore, this would be the vehicle — and the reason is structural rather than a view on the price. Roughly half of Labrador's income long term comes from a royalty: a fixed slice of the revenue produced by a mine it does not operate. A royalty is paid off the top line, so it does not care about the mine's costs, cost overruns or capital spending. That makes it a materially lower-risk way to own the same commodity than owning the miner itself.

The rest of the income comes from dividends on a small equity stake in the operation, and the whole package pays what he calls "a nice dividend."

But the conclusion is unambiguous and applies to both iron ore names: "right now we're not involved in those two." A preference is not a position.

goeasy lends to Canadians the banks turn down. The question was whether it is safe to buy after a rough stretch, and his opening line is the general principle: "lending money is easy. Getting it back is a little harder." Any lender can grow fast by relaxing standards; the bill arrives later.

His diagnosis is specific. The traditional consumer-lending business is still working. The trouble came from expanding into used-vehicle lending — a different customer, a different collateral, a different risk — and that is where the losses appeared. Meanwhile the shares now trade above book value (above the accounting value of the company's net assets), so the price no longer compensates for the uncertainty. "It's not for me, it's complicated."

Above all sits a top-down veto that would apply even to a clean lender: "I'm a little bearish on the economy so I don't want to own a lender." Lenders are geared to the economy's worst outcomes, and if you expect a weak economy the sector is the wrong place to hunt for value. He is honest that he is "not a super expert" on the name.

Propel is a US-focused subprime lender growing quickly. Tardif met management for the first time and found them genuinely convincing — "extremely enthused by the business" — and he does not dispute the growth.

He passed anyway, on a single screen that is the most transferable thing in the segment: insider selling. "We follow insider buying and selling very closely" at Timelo, and here insiders "sell regularly." Insiders sell for many innocent reasons — tax, diversification, a house — which is why the pattern matters more than any single sale. Regular, ongoing selling by the people who know most about a fast-growing lender is a signal he will not argue past, however good the meeting was.

The same economy-level caution as goeasy applies on top: he does not want to own a lender into a weakening economy at all.