The Capital Cycle Podcast - Knock on Wood

Episode Date: September 30, 2026

Two Capital Cycles for the price of one in US forestry. Edward Chancellor talks to Rob Anstey, a US Equities Portfolio Manager.For more information, or to access select articles from Marathon’s Global Investment Review publications which accompany this podcast series, please visit www.thecapitalcycle.co.uk Hosted on Acast. See acast.com/privacy for more information.

Hello, this is Edward Chancer. Welcome to another episode of the Capital Cycle podcast. I have with me, Rob Anstey, who's a US portfolio manager for Marathon Asset Management. Hi, Edward. Very nice to see you again. So, Rob, we're going to talk about an interesting asset class, which like all asset classes, goes through periods of being in and out of favour. And today, US Timberland, as we'll see, is very much out of favour. My old boss, Jeremy Grant, and writes about Timberland in his memoir, The Making of a Permabare. Jeremy's initial research conducted in the early 1990s found that US timber prices had climbed by around 3% a year in real terms over the previous 80 years, faster than the earnings growth of the S&P.

He also found that stumpage prices held up during the Great Depression. Another attractive feature of Timberland, possibly overstated, is that when lumber prices are no good, you don't have to cut it and the trees continue growing over time at no storage cost. Anyhow, back in 1990, Jeremy came across woodland in New England that was selling for less than the value of the standing timber on its property. So he bought the land and started a renewable resources group at GMO, his firm.

And over the next 15 years, timber became a hot new asset class. And early investors did very well until the arrival of the housing bust, in 2006. Now, Rob, you find that US Timberland is roughly as cheap today as it was back then. Well, yes, I'm not sure whether it's as cheap as it was then. I think if you look at the sort of stumpage value, actually, I think it's a lot cheaper now, but the land has appreciated. So, yes, you'd have to sort of look at it on an inflation-adjusted basis, but nevertheless, I think it's cheap in a global context, for sure. have two holdings in your portfolio?

Correct. I own Raynear, a sort of 8 billion market cap company, and Wirehauser, which is about 20 billion. And the interesting thing here, you know, there's very much ties in with Marathons' consolidating industries theme. Raynear has just merged with Potlatch, which in turn had bought a company called Catchmark Timber a few years ago. So essentially, Raynear is the those three Timberland reits rolled into one now. And Wirehouser has also participated in industry consolidation. So those are the only two publicly traded Timberland reits left in the US. Relatively below the radar of large cap funds. And you believe that they represent outstanding value at this current moment, in part because Timberland and Lumber, as you say, have different

associated capital cycles and both these capital cycles appear to be close to the troth. Yes, so I think it's worth thinking a little bit about association between the two. So, of course, lumber is the end product that you get when you harvest your tree. And I think there is a fascinating capital cycle going on in lumber, which we can talk about first. That's the downstream part of the equation. But then, of course, I also think there's another capital cycle. in the actual forestry, which is a longer-term capital cycle. So there's kind of two in one.

So let's start down the stream with the lumber capital cycle. Yeah, so lumber supply in North America has been declining. It's currently close to 10-year lows. Last year was the fourth consecutive year of a contraction in production. And we've met with various industry players, and one of them told us that net lumber supply growth has declined at 1.6% per annum over 20 years. So what are the reasons that lumber market has been so depressed?

I mean, there's several. You know, the obvious one most recently, of course, is that the demand side of the equation is the US housing market, which is being extraordinarily weak. Existing home sales are at around 4 million, and that's close to the lowest level in three decades, because consumers are not wanting to move off their low fixed,

rate mortgages. Yeah, and that's one of a number of bets in your portfolio that are linked to the value opportunities thrown up by a, what must turn out at some stage, to be a temporarily depressed US housing market. I think so at some point. It's the inverse of the US housing bubble in the early 2000s. And actually, I did some work at GMO on that bubble, actually for our renewable resources group. And what you found is there was massive overbuilding, of course, in the run up to the bubble. And this is the opposite. The inverse, the anti-bubble in US housing

today. I mean, I think with mortgage rates now at about 7% for 30-year mortgage rates in the US, I'm not sure when it's going to turn, but eventually it will. So the new construction is depressing. You give some numbers on the extent of the underbuilding compared to average or normalized

housing to bond. There's several estimates out there on this. So Freddie Mac estimates that since the financial crisis, the US has underbuilt around 3.7 million homes since that period. And there's other estimates out there that range between 2 and 5 million. For sure, not many industry analysts are expecting a material recovery and construction anytime soon. And there have been supply cuts, which are a necessary feature if a capital cycle is ever to recover.

Because the demand side has been so weak, you know, the lumber companies are really struggling. In the pandemic, production peaked at about 62 billion board feet, but since then, you've had six and a half billion board feet of lumber removed from the market. I said it's the fourth year of a downturn. Most producers are loss-making. Many of them have got stretched balance sheets. And how much board is needed to build house in normal times? A rough rule of thumb is that 100,000 housing starts requires a proper.

approximately 2.5 billion board feet of lumber. So if we think about the current level of housing starts, which is running at around 1.2 million, if that were ever to recover to more normalised levels, the estimate is that, you know, to meet kind of the underlying demand, the number should be more like 1.5 or 1.6 million. So in order to meet that number, it would require more lumber than is currently available from an industry that has spent the last four years dismantling its abilities to supply it. And lumber's not exactly a large input cost for US home builders. Yeah, I think this is quite interesting.

You know, lumber only represents 3% of the total construction costs of an average house, and only 2% of its selling price. And I think that's quite an amazing fact when you look at the latent value of materials in other industries. And I think Marathon, you know, we're often attracted to industries where you can find an input that's critical to production, but trivial to the customer's budget. And I think lumber sort of fits that bill. And auto class, which we discussed in a recent podcast is another example. Anyhow, there are longer term supply side structural factors, you believe, that have reduced lumber production. Yes.

And this is where I think it gets very interesting. So all of the above was a really shorter term picture. but there are some much longer term structural factors that have caused lumber production to fall. So first of all, Canada used to supply around 45% of North American lumber. That number's now fallen over a long period. It's fallen to a little over 30%, a huge reduction. And that's down to lower levels of harvesting from British Columbia, which was really caused by the destructive trail of a little creature called the Mountain

in Pine Beetle back in the 1990s. And then you've had mill closures on the back of that. And then, of course, most recently, tariffs have also affected it. And it's becoming more expensive to turn logs into boards. So this is the second thing. And I think, you know, this follows on from our earlier discussion, perhaps about container board.

So the cost of a greenfield lumber mill has gone up significantly. Today, if you were to build a 250 million board feet, sawmill, it would cost you about $200 million. So that's gone up maybe 75% since the pandemic. And that's because, you know, steel's gone up, machinery's gone up in price, labor, construction inflation. It's all hit at once. And there are actually only two equipment suppliers that make the machinery to furnish such a mill,

even if you could afford it. And the implication is that new supply, when to arrive any time soon. Yeah. I think one's always got to be careful with these kind of commodity industries. but even if you had the capital and could afford it, it would take you three years to build a mill, and that would assume that you had a suitable supply of logs. You need about two to four million trees a year nearby to sustain production,

then you need the labour and the logistics to move the logs near to your mill, because one of the things about logs is a kind of weight-to-value ratio. You know, the transportation costs may cost more than the logs are actually worth, unless they're nearby. You mentioned something that I didn't know about, which shows that there are other important sources of revenue for the sawmills that make them economically viable. This fascinates me because it shows how a lot of paper grades

and the production of paper grades have been interconnected over a long period of time. And what's happened is if you take a log, when you process that log through your sawmill, about half of it goes to making lumber. But the other half is kind of residuals. so it's bark, sawdust chips. And those residuals account for up to 20% of a mill's revenue. And they were sold to customers who would use them in other paper making grades.

So pulp. It would go to make copy of paper or writing paper or newsprint. And of course, that side of the industry, the supply has really shrunk because the demand has evaporated. So no one really wants those anymore. And is it a permanent? Is that a permanent? That's a permanent change, you see. So it's a permanent change to the market for those residuals. So what about cardboard? We talked about one of your holdings in cardboard. Another market, another market that's been depressed by the weak US housing market. Yes, so it's all interconnected. That's absolutely right. So I think that's quite interesting. But if you're the mill owner, you know, you bought your log, you're going to run that log through the sawmill. The variable processing cost is kind of fixed.

So although it may be 20% of your revenue, it actually impacts your margins much more. And that's a structural thing, I think. So that's why the industry's been struggling. And you're still bullish on lumber despite this falling demand for these extra sources of income. Can I ask you one thing which you don't mention is wood pellets? That's also a use of those residuals. And is that in this era of energy transition and so forth important? So that would be another customer of those residuals.

So let's turn upstream to your second capital cycle, namely the case for Timberland. The value of Timberland depends on where the trees are growing and what species are planted. And that varies across the US and then in other markets around the world. And it depends, you know, so in the US South, you know, it takes maybe 25 years to grow your tree and harvest it. If you're in the northwest and you're growing Douglas fir trees, they may be 40 years. They're bigger, the quality of lumber. And it's to do with the temperature, isn't it? Faster growing in warm climates.

Exactly. In Latin America, you have eucalyptus trees, which have a rotation cycle of six to eight years. The time to harvest, the quality of the wood, the size of the tree, all of those go into the sort of equation. As I was saying earlier, timber became a hot alternative asset class in the late 1990s and early 2000s when Harvard Yale and other endowments entered the timber management business. But after that capital cycle peaked around the time of the housing bust, US Timberland has been in the doldrums. Yes, absolutely. So if you look at southern log prices, they've averaged about $25 a ton for like the last 10 to 15 years. They've gone nowhere. Back in the 1990s, which, as you say, that was a sort of hot market on an inflation-adjusted basis, those prices were three times higher. So, you know, there's much higher prices in the 1990s led to an overplanting of trees.

And given the long rotation cycle, that's what's kept log prices low for the last 10 or 15 years. But you believe the bad times can't last forever? Well, I think if I'm correct on the lumber thesis and we go into sort of lumber super cycle at some point, if the thing you're producing goes up massively in price, then the input cost probably will follow as well. And particularly because those low log prices kind of disqualification. discourage planting of new trees today, right? And you believe we could possibly be close to a floor price for US Timberland? I think so because even if there were no trees on the land, you call that cutover land,

an acre might still be worth $1,500. And against current prices today? $2,000 for the US South, in the northwest, it's like $4,000 an acre. And you have a chart showing foreign comparisons to? Yes, and even in a global context, I think those US South prices look very attractive, much cheaper than the Nordics, far, far cheaper than Western Europe, not dissimilar really to Latin America, which are much lower quality, faster growth. I think we should mention here, I don't know the answer,

is to what extent is timber internationally traded? Do they bring in Latin American timber into the US market? There's some export markets, mainly from Europe, actually, but again, it goes back to that weight to value ratio. The cussar transport, these logs is going to be more than they're worth. So really you need the proximity of the trees to your sawmills. That's the key. You mentioned various sources of returns from timber investment.

Yeah, so I think there's sort of four parts to timberland returns that you can think about. The first is the earnings yield you get each year from harvesting your trees. Secondly, you've got productivity gains from silviculture. That comes in the Latin word silver, which means forest. So it's like thinning, pruning your trees, protecting them against pests and so on. The third would be higher or better usage of your land. So, you know, you can lease an acre of land for a solar farm or for carbon capture or you can sell it to a real estate developer.

And you might get anywhere from five to 15 times the acreage value. And I was also mentioning before we started this podcast how some of the US timber management organizations actually switching from forestry into agriculture, because at least they were in the middle of the last decade. Yeah, and that's the fourth component is the underlying appreciation in the actual land value. So even if there were no trees, the land rises in value, right? So what's your expected return on Timberland investment? So I think all of that, if you added up, it comes to about a sort of 8% nominal return.

That would be sort of 6% plus 2% inflation. Because you've got the inflation. The inflation comes from the appreciation, the land value, but also really the biological growth of the trees, I think. And you say that this prospect won't detain investors preoccupied with colonising Mars. Yes, I think, you know, given the sort of slow supply-side nature of timberland, you know, the over-planting trees in the 1990s, it took 25 years to appear, I don't think anyone in the US or in the equity markets at the moment is particularly preoccupied with those kind of returns. But the returns, you cite some decent historical returns. You know, there's the Encreth Timbland Index, which shows that Timberland returns from 1989 to 20203,

averaged a little over 10% a year per annum. And that's higher than the MSCI World Index, equity index of 8.8% over that time. Now, I'm a little bit skeptical when you were given numbers by that from an industry body, because obviously that period includes the 1990s, Well, they will always start at a low point. Exactly, exactly. But I think the point that they make, which I think is true, is that those returns

come with much lower volatility than equities, about half the volatility. So on a risk-adjusted basis, an 8% nominal return can be very attractive. And I think if you're buying it at a low price like you are today, that could create significant upside. And so to wrap up, let's talk to tell you a bit about Rainier and Warehouse, a brief description both those companies and why you think they're attractive? So I'll tell you what they each own. So Raynear has about four million acres of Timberland. Three quarters of that is in the south. A quarter of it's in the north-west. But it also is a top-10 lumber producer with 1.2 billion

board feet of lumber. And every $10 change in the lumber price would equate to $15 million of EBITDAF. So it's vertically integrated? Yes, for some of those mills, but obviously they own a lot more forestry. And Warehouse? Wirehouser owns 10 million acres of timberland, about two-thirds in the south, the rest in the northwest and north. It's also the second largest lumber producer in North America and the largest producer of engineered wood products further along the chain. And a $10 change in the lumber price for them would equate to $50 million in EBIT. What's their current EBITDA? So that would be on a base today of a billion. But when the lumber price spiked in 2021,

They earned $4 billion of EBITAR. So the point is that today, you know, the combined enterprise value of these two companies is $28 billion. So they're trading at $2,000 an acre with zero value for the lumber business, the engineered wood products business, higher, better usage of land, all of that, zero value for that. And finally, like other commodities, Timberland in the past has provided a good inflation hedge. and everyone is looking for inflation hedges today. The gold price has certainly done very well, and an ounce of gold now will buy you kind of two and a half acres of southern Timberland,

whereas the historic average was for a long, long period of time was 0.7 acres. So I would suggest that Timberland actually would represent a much better value inflation hedge if that's what you're looking for today. You heard it here first. Sell gold by Timberland. Thanks, Rob. Thanks, Edward. Thank you for your time today.

I hope you will listen to the next edition of the capital cycle. This communication is provided for information purposes only. Please refer to Marathon's website and the Global Investment Reviews for further information, including important disclosures.