With August’s milk production report in mind, the trading team gathered for another monthly mass balance and charting meeting. This month, though, we were blessed by two special guests: Steve Spencer and Vuko Karov from Freshagenda.
Don kicked the meeting off by modeling domestic milk production and mass balance expectations for the rest of the year, with special focus on Q4. Then, we handed the reigns over to Steve and Vuko, who guided us through a workshopping version of their Dairy Trade Simulator (DTS).
The Freshagenda team tested some ‘What if?’ questions on their model and argued that market fundamentals suggest that we should see cheese futures over $2 soon. T3 countered with some points about difficult domestic freight and contractual obligations forcing cheese from high-growth areas.
Then, to shift perspective entirely, Jacob suggested that historic correlations driving market fundamentals could break, and that there may be reason to feel bullish Class IV and bearish Class III after all.
Podcast: Play in new window | Download (Duration: 34:33 — 47.5MB)
Subscribe: Google Podcasts | RSS
T3: Welcome back to The Milk Check. This month we return to our mass balance discussion with Don Street and the rest of the trading gang, and this time we have a couple of special guests: Steve Spencer and Vuko Karov from Freshagenda, an Australia-based supply chain and market analysis firm with some great data about how milk production and pricing may evolve in 2023. Welcome to this discussion, let’s get started.
Steve: Thank you.
Vuko: Thanks so much.
Steve: Thanks for having us in your meeting. We appreciate the opportunity to join the discussion, let’s have some fun and see what it brings.
T3: That sounds great. So Don, why don’t you go ahead and lead us off.
Don: All right, here we go. Balance update, August, 2022. So in spite of Ted being more accurate than I am on these projections and winning bourbon from me, I just want to say that if USDA would get the cow numbers right the first time, I’d be much more accurate. But June was revised downwards to where it was flat. I had actually had a negative 0.02 prediction, so I think that’s reasonably close. And for Q2, which we finished down an average of 4/10 for a percent on milk. July finally goes positive. If that holds through the revision, when August is announced again, I was 2/10 of a percent over the 3/10 that was actually reported. Steve and Vuko, these are all 24 state numbers, not national numbers. I think I’m reasonably dialed in spite of the revisions, which brings us to August.
I’m, at this point, thinking we’ll be up 1% on milk, but the bottom line is that the cow herd will, in the couple of months, be higher than prior year instead of lower than prior year. Instead of being 60 or 78,000 cows below a year ago in August, when we see the numbers, we’re going to be 25 to 30,000 cows below a year ago. And in September, we’re going to more or less be equal. And then we start to see whether this grows or not. We’re going to have more cows than we did in the prior year, which will contribute to higher milk production numbers. I’ve tried to recast this a little bit to just show you the impact, because June, July, we’re up 8/10, one full percent on milk per cow, but fewer cows is the offset.
In August, I think we’ll be a little bit higher, mostly because of the poor performance of August ’21. So I think we’ll be up 1% on milk for August, but then I don’t want to say that we’re just going to continue to move higher but these changes in milk per cow are going to be 1.2, 1.3, maybe 1.4, but we’re going to be something over 1%. And then you start to add more cows. And this is of whole certainty, unless we all of a sudden see a shrink in the herd that we don’t anticipate because slaughter rates still seem to be lower, not higher. You’re going to wind up with 1.4% more milk, maybe as much as 1.8 but somewhere in that range, but it’ll be a marked difference than what we’ve experienced so far this year. Maybe just to take a look at components. And again, this is federal order numbers. There’s still really strong on fat.
We finished 21 with almost 4% fat in the milk, and it will be over 4% now, if this number holds when we finish ’22. Nonfat solids are up but protein is up more. So it’s just like having more milk production. If we look at what’s happening on the product side, which you guys know by now, I use as a proxy for demand. Class I, fluid milk, 2.3% down, totally consistent with the average of the last five years, no real surprise there. And it will just continue to chug along at about a 2% drop every month, year over year. I think the surprising thing to me is more Class II, smaller category of milk utilization. When you look at these big components of sour cream up and frozen yogurt up for the year to date, everything else is down. While ice cream is not down as much as it was earlier in the year, it’s not pushing a lot of milk demand as a category in any event.
Production changes, cheese [inaudible 00:04:50] 2.5% year to date. I think that’s pretty normal, at least from my expectations. I think the big news is always what’s the split between American and Italian, as a continuation of what we’re seeing a lot more growth in Italian, especially mozzarella. That creates more cream, probably gives the view that there’s less protein in the market, and we’ll get to that when I go to the mass balance model. Big question is when does this change? And when we can anticipate that it will change? Butter, Joe, as I told you, was actually positive in June, I didn’t think it will happen until July on production but it did, tilt down almost 3% year to date. Nonfat down 10, 11%, whether in June or the year to date, no real change in that perspective.
Again, going by class of milk. Class I, down 2%. Class II, down 2%. Cheese up two and a half, should have meant we needed 8/10% more milk, we actually only had a -4, 4/10 for the quarter. I threw in another 1% for components, which gave us a real increase in milk solids of about 6/10 of a percent. And that should have meant Class IV down just over 1%. Not exactly correct because skim non fats down 10%. Butter’s down 3%, so maybe reasonably close on the fat side but I think all this is driven by the mix of cheese that’s produced. So the more mozzarella, relatively more protein gets dried and more cream certainly comes into the market to where the butter production number is more in line with this projection.
I didn’t really do Q3, that’s a bit of a transition. I really wanted to think about Q4. Again, I think last one continues at 2% down. Last two, we could argue holiday season, maybe it will be even with last year, cheese will continue. But the big change is we’ll have positive milk production for the quarter. We’ll have still an increase in components until we don’t, so we start to look like we’re going to have a lot more milk. Thus, I think we will start to see production of skim and nonfat be higher than your prior year numbers and butter should also continue to be higher. That’s the picture as I see it on August 25th. With that, I’ll stop and see if we have any questions or disagreements.
Josh: Hey, Don. How comfortable are you in the cow numbers continuing to grow like that? It’s just so steady that you’re feeling real confident in that? Or is there any opportunity that we saw larger dairies coming online and we’ve seen some of those jumps in certain states and that that doesn’t happen as we get into the fourth quarter?
Don: At this point, I only increased the cow herd by 4,000 head a month. So from a percentage standpoint, it’s almost nothing and it won’t change the fact that we’ll still have more cows in Q4 than we did a year ago in Q4. But I don’t think it gets more bullish because the cow herd’s not going to shrink for the rest of the year.
Steve: I’ve got a question, so the recent Class IV increase was impacted by the absorption of protein into the mozzarella category, that’s likely to continue, yeah? Going forward with-
Don: I would guess it is, yes.
Steve: So we wouldn’t expect nonfat output to go up on double digits, it’d be a small increase on that trend. Is that? Yeah.
Don: I would think that would make sense.
Steve: Yeah, that’s where we have it. So yeah. I’m glad.
Don: Here where I’m showing by this simple mass balance model, I run nothing as complicated as you do, Steve. I can assure you. It won’t be up 12% but it could be up 4, 5% [inaudible 00:09:32].
Josh: Yeah. So Steve, based on some of the stuff you heard, how quickly were you guys able to punch in some numbers and gain some sensitivity analysis insights for us?
Steve: We’ve got a very similar milk production profile, Don, to what you’ve put up in terms of the go forward position. And I think Vuko, if you could share our… it’s called the DTS tools model. So it is, If you like, a what if version of the tool, it just allows the modeling of scenarios based on different perspectives on growth in production, product mix by the major players, and then looking at domestic use or domestic consumption or absorption and trade. For us, the big issues right now with China, pushing back on whole milk powder demand, New Zealand is most likely going to change its product mix, or most likely is doing a change in its product mix where it can to shift more milk to butter and skim milk powder, that’s to the tune of estimates of between 150 and 200,000 tons of whole milk powder. It’s not wanted by China in the season, we’re going to see lower demand, that’s going to push back into what you call the Class IV products. Then how does that affect a global balance? And what does that look like in total scene?
Josh: First observation is whereas on the EU side, the future’s looked aligned with your projections, your PFS, it’s not the case on the U.S. product. You’re suggesting futures are at least fundamentally undervalued a bit, so that’s number one. If we go with the methodology that Don employed that milk first flows through the class system, 1, 2, 3, and then the rest goes to four, what did you have for your projections on milk production? And then let’s adjust for Don’s.
Steve: Yeah. So as a little higher later, I think Don, we go a little lighter in August, we’re 0.8 and then we’re okay. We agree in the next two.
Don: You were even a bit more aggressive on the increase in milk than I was, except for August.
Josh: And that’s the type of insight, guys that Steve, that’s really kind of what I’m hoping to gather out of this discussion. In this case, Don was illustrating a situation that says we’re going to have more milk. We’re modeling a situation that says we were going to have slightly more than even Don anticipated. And the first observation was against the fundamental projected values you had. Our futures were already potentially slightly under price. I don’t want to get into an absolute value discussion but a sensitivity analysis so that we can see. Can you show us now with Don’s, with the adjusted numbers taking your growth rate down slightly, what that does to our turnover chart on the top left?
Steve: Back to the U.S., Vuko.
Steve: If we leave it on the U.S., let’s [inaudible 00:12:27], the move is going to come when we start to play with product mix.
Vuko: The only thing is I believe what’s happening to value is the European balance sheet is actually dominating the skim milk powder volume. In the U.S. by itself can’t really drive it that much lower.
Josh: So then that brings us to the question is our increase in milk production enough to offset the European decrease? And it doesn’t appear so, if we’re making these adjustments and it’s not having a dramatic-
Steve: No, we haven’t really blown the balance sheet apart there, Josh. There’s a lot of debate in Europe at the moment about their production being likely to be hampered by gas supply, not only is this unknown about their feed output and what it does to milk in the last two quarters of the year, but I guess the fourth quarter more so that their feed situation is pretty dire but it’s not really showing up yet in milk production numbers. So there’s a bit of a smoking gun there that might [inaudible 00:13:32] with the feed that’s in store might produce lower solids as we go later in their winter, into their colder periods and probably doesn’t get unveiled until early next year.
But the more immediate problem is that if countries have signed up to reduce their gas utilization, does food production come into the radar for that? And we see a cut in milk drying activities. The consensus is that is the case, but just trying to get the intel about how much that is. But if we assume they’re going to have a reduction in skim output below what we’ve got, and at the moment, our second half output in Europe, we’re saying it’s going to go down another 10% in the second half of the year, year on year. That’s built into this equation here. So if that gets worse, Vuko if you play with that a little bit and just take that down the second half, just take it one notch.
Don: Steve, this is on skim production, correct?
Vuko: And it takes it away from cheese, obviously. Cheese is the balancing product.
Steve: So let’s flick to the U.S. balance sheet. Now, what’s the U.S. situation? What’s that done to the-
Vuko: So you’re going below [inaudible 00:14:37].
Steve: So it’s still got our values a little higher than features, Josh, in that profile there, but the European situation, that got tighter. I think the values in Europe lifted then, Vuko, did they?
Vuko: Yeah. So one point I would make Steve is whatever we changed in U.S. and then the change in Europe, they sort of cancel each other out. So on that balance, you are looking only at an impact of about 4,000 tons by the end of H2, so not really different.
Josh: Yeah, so for the discussion because I think we can do this all day. What are the three big drivers? My takeaway is we’re going to have more milk in the U.S. Your production forecast for the EU is continued weakness year over year, down around a half a percent?
Steve: [inaudible 00:15:24].
Josh: There’s some vulnerability over the product mix in Europe going into S&P because of the gas situation and any reduction in their utilization of skim appears to be reasonably offset by U.S. availability in that simple form. So the real concerning thing about our U.S. skim price, which will contribute to Class IV is New Zealand’s season beginning with skim milk powder, so that really only emphasizes that great importance on Chinese whole milk powder demand.
Steve: It all backs up from that.
Josh: So no matter what, it goes back to if China’s buying or not, it seems like to really drive our prices. Then one final thing that I would love to take a quick peek at, if you can, in the domestic consumption side of things is just the cheese [inaudible 00:16:16]. Okay, so your projections for cheese are down through October and then flattening back out for cheese consumption-
Steve: Yeah, we’re pretty negative on that until- We are pretty flat now. I think in total terms, the recent data is slightly positive Don, is that right?
Don: Yeah. We didn’t dwell on stocks at all but they continued to inch up a bit on the U.S. side.
T3: I think the reason commercial disappearance has been decent lately has all been about exports because the domestic consumption has continued to be weak.
Steve: So in this assumption, we strip out the exports and this is just a domestic disappearance and we assume the U.S. trade share is in a different element of the model but it’s something we cater for. So we model the exports out of the U.S. as well as their domestic use.
Josh: Yeah, I was just going to ask if you could show us a slightly more negative outlook on U.S. cheese domestic consumption, how that feeds into our Class III and four pricing?
T3: I would also argue that our expectations of cheese prices as we go into 2023 are at lower levels than you’ve got there.
Josh: Those aren’t your forecasted prices, those are modeled prices.
Steve: That’s right. We study the drivers of price over time, the supply and demand variables that have the most impact on specific price series. So we look at about 21 price series globally. I think in the U.S., so we’re looking at… Vuko, what’s our series in the U.S. that [inaudible 00:17:50] largely use?
Josh: For cheeses, the blocks, skim will be in [inaudible 00:17:56].
Steve: Yeah. So what we do there, Josh, is we model the historical relationship. We use regression analysis’s, Vuko’s an economic nutrition by trade and we develop forward looking formulas based on those prior relationships, so we update that stuff every month. We need to find a tight fit between variables that drive value in the past and we use that as a basis to project forward.
Josh: Okay, so I’m going to go with the takeaway right now and then I think we can digest. But one thing that stood out to me is that you guys come up with, I would say a lot of very nice projections across total global drivers for dairy supply and demand. We have our boots on the ground in the U.S. that are quite strong in feeling some of those milk production changes, and Don does a hell of a job and modeling out what we forecast going forward.
I was a bit surprised to see… one, is that your base model was already suggesting that U.S. futures prices, at least for skim milk powder, for nonfat, were appropriately priced. Maybe you could argue under priced a bit forward but were appropriately priced. Our forecast for milk production was softer than yours so that would suggest that less flows to Class IV and the price risk could be slightly to the upside from where futures prices are now, just from a modeling perspective. Additionally, your cheese demand, I’m really curious to ask the cheese guys in our group, your cheese forecast for domestic consumption, do we think that they’ve nailed it or would we bet to one side of the equation or the other from where they were?
T3: I actually think their original projections are even a tad low, the reason I do is because I think where they have cheese price in the U.S. in the first half next year in particular, and even through the remainder of this year, too high. I don’t think we’re going to be over $2.
Steve: [inaudible 00:19:51], yeah.
T3: There you go. So I think you’re going to end up having a little bit better cheese consumption, especially in the first half of next year, primarily because you’re going to have lower cheese prices, cheese prices than $2. My expectation is we’re closer to 180, unless we end up exporting a massive amount of cheese in the next six months. Now that’s-
Josh: Just for clarity in order to get to 180, something has to change, either less cheese is consumed or more milk is produced or some other variable because that they didn’t plug that price forecast and that’s a product [inaudible 00:20:23].
T3: I get it, Josh. My argument would be is we’re going to be making more American cheese.
T3: We’ve got a new plant coming online. Everybody knows that plant’s coming online, that’s going to keep prices suppressed. So we’re going to be in the 180 region in cheese, is my opinion. We already have excess inventories of cheese right now. The only thing I don’t know is how much we’re going to export. If we end up exporting a ton of cheddar, and by a ton of cheddar, I mean more than we already are and we’re already up year over year, that’s the only way I can imagine us over $2.
Josh: So if you say the product mix within cheese promotes more cheddar, which drives the cheese price lower doesn’t that consume more cheese out of the lower price. Should we be taking that demand forecast higher?
T3: That’s exactly where I’m going, Josh. That’s why domestic consumption, I think is higher is because the lower cheese price leads to higher domestic consumption. Probably leads to higher exports too but I think we have the cheese to do it.
Josh: As a final exercise, can we pump that up a percent? And what did that do to our pricing over-
T3: It didn’t change pricing. How do you change pricing? What would cause that to happen?
Vuko: As a modeling framework, when we look historically, cheese has always been a function of skim milk powder and butter in the U.S., with the only exception is… yeah, fat and protein [inaudible 00:21:46]. And the only thing that changed that was COVID where that relationship separated. I think it’s happening currently. I saw overnight cheese blocks are 3850 on CME, but gap between Class IV and Class III looks to be persistent for the time being. So that sort of leads me to believe either cheese has to correct upwards or butter needs to [inaudible 00:22:08]. Otherwise, skim milk powder can’t really move much because it’s exposed to the European balance sheet.
Steve: [inaudible 00:22:13] add a question there Vuko, because we’ve got a slightly elevated nonfat forward value in our projections that’s lifting that cheese value.
Vuko: Exactly, yeah.
Steve: So we’re thinking the market is underplaying the tightness in the global market on skim milk powder and nonfat?
Vuko: Yeah, and we are probably exposed back to this European milk growth. I think this is where it’s at because you could be looking half a percentage point on this up or down. And on such a large milk pool of 150 billion liters plus, that’s a lot of milk. And I think the market is really uncertain on that.
Steve: So the European, we don’t see the European value flexing much but we think the tightness lives more room for nonfat in the U.S. to… if there’s greater scope in the export market because the Europeans won’t be there and New Zealand will be head to head in that respect, but there’s probably undervaluing the impact of the tightness in global availability of skim milk powder we think. And that’s impacting that cheese value, though I agree with Ted, price action in the short term looks at supply and demand and balance sheets and stocks of cheese in the U.S. And that sentiment is weaker than what the fourth value indicates. Bedrock fundamentals are still suggesting cheese is undervalued based on fat and protein.
T3: One of the arguments that I would make, and this would push back on the historical correlation between cheese and butter and powder is that where milk production is increasing in the U.S. and where cheese capacity is growing, you’re not going to have the flexibility to move that milk away from cheese into powder, even though the Class IV price is going to be, I think, three to $4, a hundred weight higher for the foreseeable future. I think you’re going to deal with the situation where you got to think of it as the milk is stuck in certain capacities because of where the new plants are being built, those contracts to supply milk to those new plants and the ability to move that milk between a Class III and a Class IV plant.
Steve: Yep, agree.
T3: And as a result of that, I think you’re going to get this break in that correlation that leads to a lower cheese price. It doesn’t fully correlate with natural supply and demand because cheese price in the U.S. is determined by the cheddar market and so it can ignore the mozzarella market when it does that.
Vuko: And one way we can actually show that is by looking at this chart, that’s why you don’t see the price move as much when you change something as U.S. domestic cheese demand, because you still have fat and protein in the background driving it.
Steve: I think Ted makes a really good point Vuko, we should be thinking about the disconnect we’re seeing now, the inflexibility in capacity. Could only really happen in the west, Ted, where there’s probably a bit more flex and then they look at a heavily stock cheese market in the center. So do they make cheese to ship east or do they actually shift a bit of milk to Class IV? That’s where there’s a bit more flicks.
T3: Well, and that’s exactly it, Steve. Because the cost of freight has gone up considerably, it’s more expensive to move milk around. You’re seeing growth in milk production in the upper Midwest, which is Class III dominated. The other place where I think you’re going to see milk production growth, especially in the next six to 12 months is going to be in Texas but in that particular situation, all the cheese plants have contractual obligations where you can’t pull the milk away from those cheese plants. So it will stay in Class III and you’ve got new production capacity, which is Class III. So the question becomes if you’re going to increase Class IV, it’s because you’re going to increase it in California. That’s a difficult discussion right now because the cost of production in California is going through the roof but it’s a Class IV dominated market and so they might be able to afford it.
Josh: Interesting. Yeah. Diego, so you don’t have to worry about milk production in the U.S. anymore, just California, New Zealand, and Europe, that’s it.
T3: We’ve been going for a while, Jake, do you have a couple of graphs you want to show us?
Jake: I can make this five minutes long and I’ll relate it to everything. So I’m going to ask a whole bunch of questions because frankly I’m not smart enough to answer them, but really tying into everything, I’m going to make the case for a neutral to bullish Class IV and a neutral to bearish Class III and really, I’ll just say cheese. Let’s hop over to the products individually, so we have cheese really on trend, more or less. We have butter way below trend. We haven’t gotten the dry products yet but we’ve nonfat way below trend. So let’s just see what’s actually going on with the pricing. I’m just going to sum this up in the form of a Class III, Class IV spread. We are really at pretty extreme levels on this Class III, Class IV spread. And I would say that spread’s going to widen out even more.
We’re on pricing support in Class III. And you have to keep in mind that this pricing support is looking at the futures. And if we go look at the Class III futures versus the CME spot market, something’s not adding up here. We’ve seen a pretty strong divergence in the Class III futures versus the theoretical Class III price based on the CME spot market. Starting about a month ago now, we have not seen the carry from the spot market to the fourth future month be this big since COVID, basically, since the pandemic hit. It’s a pretty abnormal number. So I want to take one step back and reaffirm what I’ve said. I’m going to make a bearish cheese, bullish Class IV case here, even though what our friends at Fresh Agenda said really this shouldn’t be possible based on… and I don’t want to misquote them but based on historical correlation.
And so if I’m going to make this case, I basically would need to say that there’s going to be something that breaks this historical correlation that we’ve seen. I think that a potential answer in why these correlations could be broken is just an economic event we haven’t really seen in a long time. And that would be basically asymmetric recession, where we have recessions happening around the world but at very different levels. Our 2008 recession really turned into a U.S. led, but ultimately global recession. And we’ve heard recession talked about a lot lately and I’m just curious, I’m truly asking the question of what if the U.S. doesn’t experience a recession or experiences a very minor recession, but the rest of the world does experience a recession? What is that going to do to our dairy pricing here? I think it turns it on its head and I’m going to show a few charts.
So this is corporate profits after tax in the U.S., they are insane, just insane. So just prior to the recession, we had corporate profits after tax of $2 trillion. Last quarter, it was three. I’m no math wizard but corporate profits going up literally 50% in two years, seems like a lot to me. So things are kind of looking okay here. There’s this massive battle going on in the equity markets, it’s kind of a game of chicken, I shouldn’t even call it a battle, between the federal reserve and equities. And equities are more or less guessing, they’re betting if you will that I think the fed is probably getting close to done to raising rates or they’re going to slow down dramatically. And we can see that in equity charts, this is a combination of new highs and lows. Basically, there were more new lows made every single week of 2022 up until about two weeks ago. We had our first week in the year where we had more new highs made than new lows made.
We have another one where kind of seasonally things should be trending higher here. The knife has seemed to have stopped falling on the equity side. So things are looking okay in the U.S. I could throw a lot more charts up but it does not seem like we’re going to have this dramatic blowout of the economy on the U.S. side. Now you look elsewhere and things aren’t so nice. UK inflation is expected to pass 18%. China is having property issues and they’re cutting their rates. There’s really kind of a tale of two different worlds here going on. So I can’t help but say, we really need to think about this. What if there’s more localized recessions? This isn’t a 2008. What is that going to do to our commodity prices? The other factor here, really a manifestation of these tale of two cities is going to be in the strength of the currency.
The U.S. dollar still just killing it. Really, it’s super impressive. Here it is versus the Euro. The Euro has just fallen below one U.S. dollar. This is a pretty stark chart for any Euro denominated buyer. Same thing with the Chinese currency, we’ve had this breakout that happened, the dollar broke out [inaudible 00:31:35]. So the dollar has really gained strength versus China. It really begs the question of what the heck would happen. I don’t know, as I said, I’m going to kind of be asking questions here, but it could have some material impact in breaking these historically assumed correlations that I think we might just honestly take for granted at this point, if we have a relatively strong U.S. to the world for a sustained period of time. That’ll be my Ted Talk unless there’s some things in particular that people want to see. I just wanted to kind of ask the questions and get people thinking outside, maybe what they’ve been used to for the better part of a decade here. Any requests out there?
Joe: Jake, we talked about this in the office yesterday, that butter stocks chart that you pulled up.
Joe: Can you pull up relative pricing? Because we compared it to what was it? 2014 I believe?
Jake: Yeah. Exactly.
Joe: Can you pull up what pricing looked like then?
Jake: Yeah. And it’s kind of interesting so we are way off trend in butter right now, similar to how we were in 2014. You’ll notice this below trend nature continued all the way through 2015. It was volatile in that period. We made some highs in 2014 and then at the end of 2014, we viciously came off those sides, but we weren’t right back up. It was volatile to say the least. So if we see similar volatility, this go around time will tell but you do have to see some similarity between what went on. We went from high stocks to low stocks and then we just did the same between really 2021 and 2022. Anything else? All right, that’s all I had.
T3: Okay. Sounds good. Well, Steve, Vuko really appreciate you guys joining us. Thank you very much.
Steve: Thank you, Ted. Thanks for the opportunity. Josh.
Vuko: Thank you guys.