Is the AI Bubble Ready to Pop | Anthony Scilipoti and Jimmy Connor
Summary
Anthony Scilipoti heads up Veritas Investment Research, an independent equity research firm based in Toronto, Canada.
Transcript
And if you [music] have one concern from an accounting point of view about AI and all the companies associated with it, what would it be? >> This is not a [music] question of uh you know Nvidia over capitalizing costs or or recognizing revenue too soon. I I I don't know. It's not the it's it's [music] my concern is these are in effect related party transactions because uh Nvidia is is is [music] selling to organizations and we've identified 80 of them uh that uh Nvidia [music] is selling to and so are those happening at market rates? Anthony, thank you very much for joining [music] us today. you have a very interesting background. You're a forensic accountant. What exactly is forensic accounting and why is it important for investors? >> Thanks, Jimmy. Nice to be here. Um, I think I'd like to characterize it this way. If we compare it to a an accountant and a forensic accountant. So, an accountant um studies and assesses and verifies what happened. And a forensic accountant looks at results, situations, and try and and interprets why it happened. A lot assessing perhaps blame, assessing perhaps what the drivers were that determined that outcome. >> So it sounds like you're a financial detective. >> I I'd like to think about it that way. That's a great way to think about it. >> So maybe you can just touch on some of the schemes or innovations that companies employ to deceive investors. Well, it typically comes down to obscuring economic reality. A business is performing well and the company may want to pres have investors perceive that it's performing better than that. Then the perform the performance waines and the per the company starts to use some accounting tactics to ensure that the results continue to look as though the business is doing well. And what that putting it simply is it's when you're recognizing revenue ahead of the economic reality of earning it or recognizing expenses later than their uh economic reality of when they're incurred. >> I want to get your views on what's happening right now with artificial intelligence and valuations. But before we do that, I want to go back in time to the tech bubble. And one of the market darlings that that time was Nortell. It was spun out from BCE. The stock peaked in July of 2000 at $125. And one year later, in July of 2001, the stock lost 94% of its market value. And there was many reasons why. A lot of it had to do with what was happening within the overall market at the time. But they were also employing some deceptive accounting practices. Maybe you can just touch on that because you were you did a research report on Nortell at that time and I believe you came out with a silver report on it. Is that correct? >> That's right. First work we did was in February of 2000 and you know I would say Nortell is interesting. Were they doing deceptive practices? Well, that only came up later. We in in other words, you can only tell if a company is uh obscuring the numbers if you know what's actually happening behind the scenes. As outsiders, as those of us looking at publicly available information, the only thing we can see is where there's changes, where the company uh doesn't provide enough information to make a conclusion, where the company is employing a or is presenting their results as I alluded to earlier where they tend to um diverge from what you see happening in the economic reality. Let me give an example of what we meant and what we looked at back in at Nortell days. So put simply uh Nortell was in essence fueling the uh the market uh for um uh the internet. Okay. And its customers some of which we've talked about getting ready for this 360 networks JDS Unifase etc. the startups of the day that were building out the internet. Well, they were they didn't have funding. So, Nortell would sell them product and they would in turn u say enter into agreement with Nortell, listen, we can't pay you in 30 days. We're going to pay you in 60 in 90 in a year. And so then Nortell would have what's called a long-term receivable or the company would say, "Listen, I'm going to spend 10 million with you, but I need another 5 million to buy things from Cisco, Lucent, Microsoft, etc." And so they would get a line of credit from Nortell. And so these in these um other types of assets, let's call them long-term receivables and those longerterm loans, working working capital or lines of credit, those would show up as long-term assets, that would then not be considered as part of working capital. So when analysts would look at the balance sheet of Nortell, it would look like it was very liquid. it wasn't a current asset that was being uh inflated by this long-term lease loan. It would be long-term assets and liabilities would be fine. So, it was then on the cash flow statement, you saw that uh a normal receivable would show up as working capital, but a long-term receivable would show up as part of financing items as would the work the the line of credit. So, when analysts would go and look at the uh cash flow of of Nortell, well, operating cash flow looked great. free cash flow looked great because that number less whatever capex they had would show a good a decent number and the and they would look at that looking at their Bloombergs and fact sets because everybody wants a quick answer and they'd say that's fine but that misses the economic reality they were not they were selling on long-term and offering credit and so what we did was we looked at that detail and say well if you're offering that type of uh product and that's how you're conducting your business how is that being reflected in your financial statements well that's a long-term receivable. So that's showing in financing. So that should be part of free cash flow. And once you included that along with the loan, the the working capital loans that Nortell was offering, you could see that the company's cash flows were diverging materially from their earnings. And that all came to roost really in 2001 when the company ended up issuing debt and then basically the wheels fell off. >> And this whole process is known as vendor financing. And it wasn't just Nordell, it was Cisco, Seammens, Lucent, so many other companies were doing it. >> And and you know when you think about it, they were they had launched a new industry, a new paradigm. The internet was born. You needed to have servers. You needed to have this optical fiber cable that Nortell had created. And it needed funding. So there's only so much you could get from investors. the participants, those with the deeper pockets, Nortell Cisco and so on, they provided that uh to help sort of fuel the growth of the industry. >> I forgot about fiber optics. That reminds me of another company at that time was Global Crossing. Do you remember that one? >> Of course. >> So, Anthony, let's bring it ahead 25 years to the present day. And so, now we're living through some very unusual times, fascinating times with artificial intelligence. And when OpenAI first released Chat GPT in November of 2022, OpenAI was valued at $20 billion. And here we are just a few years later, it's now valued at $500 billion based on the last raise. Also in November of 2022, Nvidia, the other poster child for all things AI, it was valued at 400 billion. Now it's valued at over $4 trillion. And one of the things we keep hearing about this whole AI thing is circular financing. So maybe you can just talk about that. What exactly is circular financing? How does it compare to the vendor financing we saw back in the early 2000s? >> Okay. So we've discussed vendor financing. Circular financing or or so it's now been called. Um and and this is something we've been writing about for the last couple of years. In essence what's happening is similarly uh Nvidia is uh financing or is fueling uh a new paradigm let's call that AI artificial intelligence and so in order to do that you need funding so what Nvidia is doing is it is investing in startups andor mature organizations that are um that need to purchase its chips to build out their infrastructure and create the large language model simulation uh systems. And so in order to do that, Nvidia invests, let's say, $1 or one, let's call it, you know, 10 million and then they'll get a number of other investors to invest another 90 million and you now have a hund00 million investment in this startup organization. Let's call it Open AI. OpenAI then turns around and purchases chips from Nvidia. So that becomes circular. I give you one, you turn around and buy products from me. Why is this interesting from an e that's that's the economic reality of what's happening? But on from a financial statement perspective, what's going on? Nvidia takes the 10 million, it turns that cash into a long-term investment. Okay. Now this investment could turn out to be fantastic because that company as you mentioned let's say take the case of open AI it went from 20 billion to 400 billion so on Nvidia's finan on Nvidia's investment financial statements they're able to book that increase in valuation so they're their shareholders are benefiting from Nvidia making this seemingly great investment and this continues but from now the financial statements they have this investment They record revenue. The products get sold. Once they're sold to an to a business in which Nvidia is invested in, are is that occurring at as an armlength transaction? Is that occurring at market rates? Do you think that those companies that Nvidia's invested in? Do they care what they're paying for the product when they stand to lose perhaps some more investment from them? And so they take that that uh in that investment. Okay. So now it's sitting on on Nvidia's balance sheet and there's no cost associated with that investment. This is different than what Nortell was doing. Nortell would in essence sell product and offer you time to pay for it. Nvidia is not doing that. They're they're they're making an investment in your business and you're turning around and purchasing product. >> So let's talk about Microsoft. Now they have a legacy software business but they also are building out these data centers and uh they own 27% of open AI and so when you look at these companies Microsoft is a publicly traded company but where do you see the risk there with this scenario because it seems to me when you look at open AI it has revenues of $10 billion but it's got financial commitments for $1.4 trillion. Okay. and they still have not provided a game plan on how they're going to provide for those commitments. Now, maybe their revenues are going to double or triple here in the coming years, but that remains to be seen. But once again, part of your job as a forensic accountant is identifying risk. Where do you see the risk in this scenario? >> The risk is that any business is only as strong as its customers. So if Nvidia, Microsoft is investing in startup organizations, um they're obviously banking on profits to occur later on. Now, because they're p they're those customer those those invest positions in also become their customers, they don't actually care if they don't make any money on their investment. But from the standpoint of the overall industry, what's what this what the weakness and the risk that I see is in order for open AI to continue to to to work andor perplexity, anthropic and so on, they need to eventually convince and investors that there's a view to a profit. And the longer that time frame extends, perhaps investors will become impatient. And there in lies the risk because if they become impatient then Nvidia's customers no longer buy product from them and then Nvidia's growth slows and then it circulates the other way. There's no investments. There's no market cap expansion. Margins contract >> and that's basically the same thing we saw happen back in the early 2000s when we had a pullback in the market. People just stopped spending. Nortell wasn't getting paid. No. Nobody was buying their product. >> The customers had to pay back that purchase that they made from Nortell. That means they needed to get capital from outside sources. If the moment that that capital dries up, the whole circle dries up. It always rests ultimately on liquidity. Liquidity is like oxygen, right? You don't know how much you need it until you don't have it. And right now Nvidia is fueling that with liquid its own liquidity as is Microsoft and benefiting as a result. But if that dries up from other sources, that's going to be a real problem. >> So you raised a very interesting point there, the importance of liquidity. And another hyperscaler that's coming under pressure right now is Oracle. It's down 40% from its top that we just saw here in the last couple of months. Y >> and one of the issues is these commitments it has with open AI. And one of the things that's come out here in the last few months is that the market is getting concerned about being able to monetize AI. And the hope was that when they build out these data centers, they're going to monetize AI with by 2027 2028. Now there's speculation it might not happen until 2030, which really impacts the whole profitability of these companies like Oracle. And the other interesting thing about Oracle is that their credit default swaps have really blown out here in the last couple of months. 125 to 130 uh highest it's been since 2009. But maybe you can just speak to that because once again it goes back to what you were saying. Everything is working until it doesn't and all it takes is one little crack in the foundation and income's falling down. you know, investors when they're investing in uh growth companies, they're buying dreams. And those dreams are based on stories. And if that story is look, we're going to be profitable eventually, we're going to change the world, and as long as that is believed, then everything's fine. It's when not when the providers of that story or the the organizations and the executives that underpin that story when they change their story as you said or um with open AAI and and Oracle having to change and extend out that uh view to a profit as I mentioned right I said earlier I said it depends on if if investors lose patience well if investors lose patience then we start having a problem and you got to following a bit the tea leaves, right? Everything matters. If we go back a few quarters, no one was asking about inventory turnover, AR turnover, changes in depreciation. These are things we've been talking about for the last couple years related to the players in this industry. And yet, no one wanted to read about it. No one called us. No one cared because the market was going higher. and price, okay, creates uh narrative. If the price is going up, I'll just justify by saying I'm buying a dream. The moment the price starts getting choppy, now people start asking questions. Wait a minute, when's that when's that earnings coming? When's that cash flow coming? What's going on with that inventory? What's happening with your margins? If that start the narratives start changing into that, then the story is changed. And when the story changes, then the multiple may have to be impacted. >> And you touched on depreciation, and this is another concern we've uh seen in the last few months. Michael Bur came out with a report and he's saying the depreciation associated with these chips. Uh it's too long. And he's basically saying the life of these chips because of technology is changing so fast that the life of these chips is much smaller. Yeah. Maybe you can just speak to that and how that's going to impact. >> I I think this is a this is an area of uh significant debate that only time will tell and there's a number couple of different factors that I would say. So yes, you know, Nvidia wants you to believe that they're short because they want their customers and the market investors to believe that you need to have the next best thing. But if you travel across the pond to the other side and look at what's happening in China and I recently visited um they created deepseek which uses a different algorithm to do its uh um analyses and as a result uses less power and works on old older dated chips from Nvidia. And I I haven't tested the speed, so maybe it's a little bit slower than than Chad GPT and and Perplexity and so on. But in when we're talking about something that answers you in seconds, if it takes a few seconds longer at this point, I'm not so sure that people are too worried. Um because there's going to be different uses for AI. Some that require instant answer and some that may not. So I think there's still debate. uh there is definitely the question that they do get burned out especially over high use that they've shown that they they lose their efficacy but at the same time I think there are ways that will extend out their life um and that's not a good story for Nvidia because if no one needs to invest in the new black well to get to the next who knows what then where does the next growth come from >> and if you have one concern from an accounting point of view about AI and all the companies associated with it, what would it be? >> This is not a question of uh you know Nvidia over capitalizing costs or or recognizing revenue too soon. I I I don't know. It's not the it's it's my concern is these are in effect related party transactions because uh Nvidia is is is selling to organizations and we've identified 80 of them uh that uh Nvidia is selling to and so are those happening at market rates and um if they are fine but if they are not uh then and and those organizations cannot get external financing um then Nvidia may have to sell to other organizations at different prices and that could affect its its margins. We don't know you you with with accounting or or when you're looking at financial results you end up with questions. You end up with looking at trends and situations that don't m don't match well with the economic reality. I hope that's coming through in what I'm talking about. And so then you continue to ask the question why. Remember I said what's the what does a forensic accountant do? It continually asks why. I'm a I'm actually a pain in the ass. When people tell me things, I always ask why because nothing happens without a reason. It's important for Nvidia and all the other major players to continue to invest in their customers because they need the AI to fuel the liquidity. It also gives a once you have the the the you can say that Nvidia or Microsoft or some other large mega company uh MAG7 company is invested in you well you use that as a badge of honor don't you want to invest and so again when the narrative changes that's what makes me nervous it's not something that I can pinpoint specifically this account on on Nvidia's financial statement makes me concerned >> so we started this conversation by discussing Nortell which was a Canadian darling back in the early 2000s. >> I want to ask you about the Canadian Telos BCE, Telus, Rogers and these companies peaked in 2022. Y >> and since that time both BCE and Telus are down 50% give or take. Yeah. >> What is happening within this sector? Because at one time this was the sector you would want to own for stability and growth, dividend yield. >> Yeah. In many ways they're a utility. They make an investment in infrastructure like a pipeline in or a or a power generation. In this case, they put in uh you know uh cellular nodes and they put in uh uh you know infrastructure cabling and so on and then they sell use time on that. And um this is actually something that we warned about. I mean we were first to warn in 2023 that there were cash flow problems at BCE. What I think has happened is uh the whole industry has in essence done it to itself. So you have a government that needs revenues and you have seemingly an insatiable demand from customers for internet capacity and mobility on both mobile and and on landline, right? And so what do you have? Well, we have 3G. We got to go to 4G. We got to go to 5G. Well, who sells that? Well, the government. government sells the space on the f in the um on the networks basically the the the data the waves and so the government there's no price for this it said hey I have this much of airwave available in 5G network and so the three major players beat themselves up to bid for that and paid a fee to the government and then the government says to them oh by the way now you paid for this, but I'm watching how much you're charging customers. Okay, you're going to have you're charging too much to customers. You're going to have This is not good. We're watching, right? And there's all this pressure on them not to increase prices, but it doesn't make sense. When they let Enbridge put in a pipeline or they get they let um Hydro1 build a a power generation facility, they allow them a a rate of return commensurate with their cost of capital. Why is it that they don't do the same for Telos? And so because they're under pressure, can't charge what is perhaps a return that they should charge to make money because Canada, you know, they say, "Well, Canada pays too much in mobility." But, you know, when I'm up in North Bay and I go to Sudbury and I go to Timmans and I turn on my phone and it works and it's fast, that costs money because there's very few people there. And you know, it's easy to put up cellular satellite sites here in in dense populous areas, but Canada's spread out so wide that that's where the cost comes in. Anyway, maybe I'm sounding like I'm promoting their businesses, but it's it's never a simple answer. And so that's that's how I think we got into the problem. Um, for the companies themselves, they overspent. Okay? And it's a it's a business that's always under pressure because we got three main players fighting against themselves largely on price. And so uh we saw that their cash flows were eroding. Everyone was buying them comparing them to the companies I just mentioned Hydro1 and Embridge for example. Well, if they're pay out paying out this level of yield and they're increasing their dividend each year, if I want to be considered as a comparable on the dividend funds and for everybody looking for dividends, I also have to do the same. And so, they were doing the same. But then it got to a point where they could no longer mask the decline in their free cash and they ended up were pay paying too much in dividend. And here we are. Rogers different because they didn't uh um sort of build out as much. They they made the big acquisition of Shaw and so that took added debt for them and they didn't increase their dividend as much. Their dividend yield always stayed lower. So I think that's what positioned Rogers in a better better from a cash flow perspective. Um it's been the one that we we like best. Uh and as well Quebecor, they bought up the the freedom assets, the mobility assets from from Shaw. Very good move for them. They took on that which generated significant cash flows, opened up a whole market to them. And there you go. >> The dividend yield on both BC and Telus is still relatively high. I believe it's eight or nine% and that's always a good reflection of that there's still trouble. >> You know, I I I have to make a plug here. I I I wish investors would stop aggressively, I'm saying this, using payout ratios tied to earnings. Earnings is made up. It's an it's an it's a it's a subjective number. Cash flow also has its subjectivities, but at least I can it it starts with and moves around actual cash flow. Dividends have to be paid out of cash flow. So the dividend when you calculate payout ratios what we do while it's amusing to look at the payout ratio r related to earnings we look at it tied to cash flows and that is where you'd see that the c payout ratio okay is too high more than 100% of the free cash they're generating and then you have to normalize the free cash because there's no textbook way of calculating free cash flow so each BCE Rogers and Telus calculate it differently. So we normalize those in order to come to a number such that we can say which one is more at risk than the other. >> The other thing I remember about BC in 2025 they closed on the Maple Leaf Sports deal. Uh they sold 37.5% to Rogers. Rogers now owns 75% of it. And in my mind like I'm thinking why the hell did you do that? Because if you want to own anything in this environment we live in now, it's sports content, right? And you look at Netflix, one of the greatest media companies in the last 10 years, what are they doing? They're going all in on live sports. They're paying $500 million a year for wrestling. They just paid $150 million for two NFL games on Christmas Day. And yet BCE just sold their stake in Maple Leaf Sports. What are your thoughts on that? You know, that's why debt kills. When when individuals and organizations overlever themselves, they can no longer make business the best business decisions. They have to make decisions to solve for a problem. And that's when uh there's opportunity obviously always for the buyer, but that's when you that that's why companies that are highly levered are higher risk. Anthony, I want to get your views on the Canadian economy and how it relates to the real estate market, especially here in the greater Toronto area. And I guess you could characterize the Canadian economy is weak. And the unemployment rate across the country, it's around 7%. In the province of Ontario, it's 8%. In the city of Toronto, it's 9%. And this trend has been going up for many months, many quarters now. And what does all of this mean for the real estate market across the country and more specifically just in the Toronto area? >> Let's break that down a little bit. I want to start with your comment that the economy is weak. Well, GDP is still growing uh >> moderately >> moderately but growing. Um the retail sales are weakening but still holding up. Um yes, we have seen an increase in unemployment which typically is a sign of weakness. But the problem with uh any metric is it's just a ratio. We've spent an inordinate amount of time focused on this in particular and um our analyst Shallab Guard's done a great job gone back to and this is tied to the real estate conference which we've been having for the last 13 years. We actually publish on a quarterly basis just a report specifically on real estate. So, we went back to 81 because that's when uh the great the greater Vancouver area uh rolled over and we went back to 91 because that's when the GTA rolled over. And we looked at what happened. Remember, we're always in the question asking why. Everybody thinks it was unemployment. Well, unemployment was already rising prices and prices continued to uh increase during those time periods. What happened ultimately was it was not until there were job losses. Job loss means Joe has a s has a job on in January in February he doesn't have a job. March he has no cash flow. Now we got a problem. Unemployment is a ratio of the total number of humans that are looking for a job over the total number of humans of age and ability to have a job. If we had under the previous guy that was sitting in Ottawa increasing our our num our numbers of immigration, well, we brought in many individuals and the number of individuals that came in was more than the jobs that were being created. So, we had unemployment, but those individuals didn't have a mortgage because they just showed up lately. Really hard for them to get a loan. They didn't have significant consumer debt because they didn't have the cred the credit quality to get it. And so, while the number on the surface goes, we're flashing, we got a problem with unemployment, when you look underneath, there aren't yet job losses. Now, they've just started. There were some job losses in July and August. And my concern, which ties to everything you've just said, is that real estate is now getting into a headwind. And that headwind is there's no new housing starts. When a building is finished, if there's no new building to go up, what do you do with the plumber, the electrician, the drywall guy, the form guy, the skipper who was on the job? They go from doing something to not. You can't build half a building. You can't do half a window. You'd build a whole window or no window. And so we have no new projects coming up coming some part in the middle of 26 and onward. That means there's going to be a huge headwind if we don't get infrastructure spending or something else for these individuals that are in the construction industry to do. And my fear is we have like 10% of our economy, 10% of our employment in construction related activities and that only includes those that report their income. Now I know this is not something you, your family, anybody in your area would do, but there are parts of this country that when they hire individuals and organizations to do their work in construction, don't ask for a bill. Those guys aren't being reported in that income number. And I hear stories that that number could be very significant. If that slows, you see a further erosion in the real economy and that's when I think there'll be a problem. So for now, we haven't seen any real problems. Mortgages have been coming up for renewal at higher rates. How is it possible people have been able to make it work? They still have a job. >> Very good points. And uh just to provide some more context, in 2025 over 1 million mortgages came up for renewal. And in 2026, I believe the number is 1.2 million. All right. So the Bank of Canada has been very aggressive in trying to get these rates down so people can finance these mortgages. >> And I'm sure they're very concerned about it because, you know, if your mortgage payment is $2,000 a month and then it goes to $4,000 a month, you got an issue. And to your point, if you lose your job, then you got a serious problem. And going back to unemployment, we still have this trade war going on with the US or >> it's not resolved. Let's put it that way. >> And we're seeing a lot of negative impacts. >> Stalantis just shut down a plant in Bmpton, 3,000 jobs gone. Then you look at the suppliers that feed that one plant, there's probably 3,000 more. General Motors closed down another plant where they were making that Bright Drop EV uh delivery van. That's another thousand jobs gone. Al Steel just laid off a thousand people, one-third of their workforce. So there's this big trickle down effect. And so when you look out into 2026 or later in 2026, where do you see the unemployment number going? And how does that impact the real estate market? >> Well, I I think the unemployment rate will rise and it'll rise from job losses. which is the danger which has the trickle down as you described to the overall economy. And the other interesting part about our current economic situation and this comes up in the US as well is it's a K-shaped economy. So think of the shape of a K. So those with assets since since uh COVID have done phenomenally well because the stock market's up three-fold. Uh their their house went up in value. They're doing great. Those that are living on paycheck to paycheck, I call that Joe six-pack. They're suffering because inflation is the death smell to an economy because what it and it affects everyone whether you're rich or you're poor. You show up and you see that the h that the the the stake used to be 20 10 bucks 20 bucks and it's 40 bucks. You say whoa even if it's not a big material month amount to you say whoa maybe I'll buy this small cheaper cut. And so what's happening now is if those on the bottom part of the K and and tied to the inflation, what happens is, you know, inflation they say, "Well, inflation is down. We're at 2%, Anthony. Now, isn't everything fine?" Well, it's a rate of change. So if year one we went up 8% and the next year we went up 6%. And now we're down at 2%. That feels good. But that forgets the point that we were 8 plus 6 plus 2 plus the amount the accumulation of that we're probably close to 20% increase in price to that human from 3 years ago. And so you see this we're at the latter part of the stage of the shift of the economy. It just came up in Air Transact. Why do you think that the the the pilots were willing to not work and they wanted and eventually got like a 60ome percent increase in salary? Because that's what they needed. They hadn't had a a a wage increase in years because they were a onesie, twozie, threes percent because that's what the inflation rate was at. That seemed to make sense. Now all of a sudden it no longer makes sense. And this I think is going to be the strain. the fact there's there's there's this strain between the worker and the company. The company willing to shut down all its airlines, its it aircraft and the and the and the workers willing to not work in order to uh because they're both at an impass. Now we're we're we're at that point where impasses are happening. That's when the real cracks come. I think we both agree the economy is weakening and we're seeing these unemployment numbers tick higher. But what's really surprising to me is when you look at the TSX, it's up 30% on the year. Just an amazing year. S&P still up 15%. So massive outperformance in Toronto or in Canada over the US, but uh a large component of the TSX are financials. I believe it's 30 to 35%. And there's been no weakness in these Canadian banks at all. Does that surprise you? Um no uh first answer and and secondly I think it's really important you know that people uh need to stop looking at the index be it S&P TSX uh the NASDAQ and go well if the Dow or the Dow the Dow's up 3% everything must be fine in the economy everybody must be doing fine or the S&P's up and therefore or everything's fine or the S&P's down and therefore everything must be bad. It's just a metric. 500 companies and that's how their valuation changed that day, that year, that month. It may have absolutely no uh link coincident or otherwise with what's going on in the underlying economy. the market is looking forward and it contains companies that may not be a reflection of the over outlying economy, right? Like real estate as a percentage of the Canadian index is virtually zero. We don't have any public real estate companies. So we're having a struggle in real estate in in real estate, but there's nothing to be affected in the index. Well, that's because there's no real estate companies. Do we have REITs that own real estate? Okay. Some have done decently well, some have not, but we don't have like constru construction companies uh that that build things on our index. Okay. And so that that's what I think creates this this uh dichotomy that you need to that investors need to be careful of. >> And so when you look at the financial sector overall, you're not concerned about the Canadian banks? Well, I listen, I'm I'm I'm concerned because their their valuations are at that the more uh they're elevated relative to historical averages. Um but uh I I'm I would be I'm getting I get more concerned as uh job losses occur. And I'm not we're not concerned because of mortgages per se, banks. Um you know, I I'd be I'm much more concerned about private markets. I'm very concerned about the private mix. I'm concerned about all private credit. I think that's where the the biggest risk and bubble lies. Um, but where I look at uh for for the banks themselves, it they'll lose money if they if it if we have problems in credit cards in unsecured lines of credit, those banks that have outsized exposure to those sectors of the consumer or those type of loans to the consumer are going to be more affected than those that don't. >> So overall, when you look at the Canadian economy, we we talked about the Telos. Do you still have some concerns there associated with a couple of the big players BC Intel? >> Oh, I would say I would say there could be there could be opportunity at some point like we're really getting washed out right we're we're down as you said significantly now it's going to be is is their business going to stabilize? I think investors once they see their business stabilizing, they'll be there'll be some turn uh because things have really gotten beat up here. And I think the government is more interested in other things other than your credit your uh mobility bill. They're going to be having to worry about unemployment such and so they're not going to be breathing down the big three's neck saying, "Hey, you better cut costs." >> Are there any other sectors within the Canadian economy that concerns you where you see a risk to investors? No, I think you you you touched on on on banks. I think the private space uh any any private equity, private credit, those are those are space that I would be concerned about. Um I would say uh consumer uh discretionary uh is one as well. We don't have a lot of that in Canada. Uh but those spaces are ones where we're going to see an issue because uh if we do end up seeing the job losses materialize uh then people just spend less you feel and then you have this the wealth effect in reverse even if you have assets and you have money you just think twice it just happens and so that I think is what often gets missed when economists look at numbers they just say well everything's fine or everything's not fine well you need to go on the And that's really something we do a lot of is actually go and shop, actually go and visit what a condo site looks like and what's happening. That's where we get some of our best information. >> Okay. So, you do a lot of work on condos within the greater Toronto area. What are your thoughts there? Because everything I read, it's a total collapse. >> Yeah, it's it's uh it's really bad. Um the the problem is and it's a slowm moving uh train wreck because you have uh many uh properties that have been purchased at high prices. They're diff they can't get the financing to close. they've uh overextended themselves many now there's too much supply and those can't they're not renting because at the price that the that the landlord wants because in order to make the numbers work they need a higher rent but but that is not what the what the what the uh tenant is willing to pay. So you have this impass among renters. You have the impass among buyers because the buyers think that the price should be lower. The sellers still believe looking backwards that the price is worth more. So you have this stage which I call denial. And so we're in the still the stage of denial. And there's five, it's like grief. There's five stages. The first stage is denial, right? Then you have anger. You call your agent. You get mad at your kids. your mortgage broker, your banker, everybody did something wrong, that's why it didn't sell, right? Then you end up getting depressed, right? And you you end up uh ultimately then calling back the guy that gave you the low bid and he tells you, "Look, I told you I was going to pay you a million. Now I'm willing to pay 800." And then the worst case, you fall into acceptance at the end and you end up buckling. And I my advice to people is, you know, if you don't need the property and it's not generating cash and it wasn't generating cash in the past, let's just get rid of it. Don't wait. If you if the cash doesn't matter to you, you can hold through the through the end. The the mortgage payments are not a big deal for you. Do the worst case scenario. It's not rented for a year. Can you still hold on? Yes. Okay. If the good location, then let's not worry about it. But if the location is not good, then start thinking about uh making different decisions. >> Well, Anthony, this has been a great discussion. I want to thank you very much for spending time with us today and sharing your thoughts on forensic accounting and the importance of it to investors. If somebody would like to learn more about you and your firm, Veritas, and the services that it offers, where can they go? >> Veritascp.com. That's the base. We're an independent equity research firm um covering mainly Canadian companies but we also do some some uh US companies and uh then we launched an asset management firm called Veritas asset management. We take those ideas and package them into portfolios. And then we launched a phil philanthropic uh platform where we offer donor advised funds and we do research on charities so that when donors make donations, they don't just give with their heart but they give with their with their mind as well and are armed with research. >> And Anthony, you also have a podcast. >> Yes. Who wants to listen to it? >> Yeah, our my podcast is called the factfinders. We started that uh in uh March, April of 2020 and it was really uh again my uh search for why uh the to go on the ground talk to industry professionals about what's going on in the economy. Uh and so our mission here is to help investors make better investment decisions. So we'll have different experts uh on the on the on our platform and uh you know try to offer that to our investors. >> Well, this has been a great chat once again. Thank you. >> Thanks for having me, Jimmy.
Is the AI Bubble Ready to Pop | Anthony Scilipoti and Jimmy Connor
Summary
Anthony Scilipoti heads up Veritas Investment Research, an independent equity research firm based in Toronto, Canada.Transcript
And if you [music] have one concern from an accounting point of view about AI and all the companies associated with it, what would it be? >> This is not a [music] question of uh you know Nvidia over capitalizing costs or or recognizing revenue too soon. I I I don't know. It's not the it's it's [music] my concern is these are in effect related party transactions because uh Nvidia is is is [music] selling to organizations and we've identified 80 of them uh that uh Nvidia [music] is selling to and so are those happening at market rates? Anthony, thank you very much for joining [music] us today. you have a very interesting background. You're a forensic accountant. What exactly is forensic accounting and why is it important for investors? >> Thanks, Jimmy. Nice to be here. Um, I think I'd like to characterize it this way. If we compare it to a an accountant and a forensic accountant. So, an accountant um studies and assesses and verifies what happened. And a forensic accountant looks at results, situations, and try and and interprets why it happened. A lot assessing perhaps blame, assessing perhaps what the drivers were that determined that outcome. >> So it sounds like you're a financial detective. >> I I'd like to think about it that way. That's a great way to think about it. >> So maybe you can just touch on some of the schemes or innovations that companies employ to deceive investors. Well, it typically comes down to obscuring economic reality. A business is performing well and the company may want to pres have investors perceive that it's performing better than that. Then the perform the performance waines and the per the company starts to use some accounting tactics to ensure that the results continue to look as though the business is doing well. And what that putting it simply is it's when you're recognizing revenue ahead of the economic reality of earning it or recognizing expenses later than their uh economic reality of when they're incurred. >> I want to get your views on what's happening right now with artificial intelligence and valuations. But before we do that, I want to go back in time to the tech bubble. And one of the market darlings that that time was Nortell. It was spun out from BCE. The stock peaked in July of 2000 at $125. And one year later, in July of 2001, the stock lost 94% of its market value. And there was many reasons why. A lot of it had to do with what was happening within the overall market at the time. But they were also employing some deceptive accounting practices. Maybe you can just touch on that because you were you did a research report on Nortell at that time and I believe you came out with a silver report on it. Is that correct? >> That's right. First work we did was in February of 2000 and you know I would say Nortell is interesting. Were they doing deceptive practices? Well, that only came up later. We in in other words, you can only tell if a company is uh obscuring the numbers if you know what's actually happening behind the scenes. As outsiders, as those of us looking at publicly available information, the only thing we can see is where there's changes, where the company uh doesn't provide enough information to make a conclusion, where the company is employing a or is presenting their results as I alluded to earlier where they tend to um diverge from what you see happening in the economic reality. Let me give an example of what we meant and what we looked at back in at Nortell days. So put simply uh Nortell was in essence fueling the uh the market uh for um uh the internet. Okay. And its customers some of which we've talked about getting ready for this 360 networks JDS Unifase etc. the startups of the day that were building out the internet. Well, they were they didn't have funding. So, Nortell would sell them product and they would in turn u say enter into agreement with Nortell, listen, we can't pay you in 30 days. We're going to pay you in 60 in 90 in a year. And so then Nortell would have what's called a long-term receivable or the company would say, "Listen, I'm going to spend 10 million with you, but I need another 5 million to buy things from Cisco, Lucent, Microsoft, etc." And so they would get a line of credit from Nortell. And so these in these um other types of assets, let's call them long-term receivables and those longerterm loans, working working capital or lines of credit, those would show up as long-term assets, that would then not be considered as part of working capital. So when analysts would look at the balance sheet of Nortell, it would look like it was very liquid. it wasn't a current asset that was being uh inflated by this long-term lease loan. It would be long-term assets and liabilities would be fine. So, it was then on the cash flow statement, you saw that uh a normal receivable would show up as working capital, but a long-term receivable would show up as part of financing items as would the work the the line of credit. So, when analysts would go and look at the uh cash flow of of Nortell, well, operating cash flow looked great. free cash flow looked great because that number less whatever capex they had would show a good a decent number and the and they would look at that looking at their Bloombergs and fact sets because everybody wants a quick answer and they'd say that's fine but that misses the economic reality they were not they were selling on long-term and offering credit and so what we did was we looked at that detail and say well if you're offering that type of uh product and that's how you're conducting your business how is that being reflected in your financial statements well that's a long-term receivable. So that's showing in financing. So that should be part of free cash flow. And once you included that along with the loan, the the working capital loans that Nortell was offering, you could see that the company's cash flows were diverging materially from their earnings. And that all came to roost really in 2001 when the company ended up issuing debt and then basically the wheels fell off. >> And this whole process is known as vendor financing. And it wasn't just Nordell, it was Cisco, Seammens, Lucent, so many other companies were doing it. >> And and you know when you think about it, they were they had launched a new industry, a new paradigm. The internet was born. You needed to have servers. You needed to have this optical fiber cable that Nortell had created. And it needed funding. So there's only so much you could get from investors. the participants, those with the deeper pockets, Nortell Cisco and so on, they provided that uh to help sort of fuel the growth of the industry. >> I forgot about fiber optics. That reminds me of another company at that time was Global Crossing. Do you remember that one? >> Of course. >> So, Anthony, let's bring it ahead 25 years to the present day. And so, now we're living through some very unusual times, fascinating times with artificial intelligence. And when OpenAI first released Chat GPT in November of 2022, OpenAI was valued at $20 billion. And here we are just a few years later, it's now valued at $500 billion based on the last raise. Also in November of 2022, Nvidia, the other poster child for all things AI, it was valued at 400 billion. Now it's valued at over $4 trillion. And one of the things we keep hearing about this whole AI thing is circular financing. So maybe you can just talk about that. What exactly is circular financing? How does it compare to the vendor financing we saw back in the early 2000s? >> Okay. So we've discussed vendor financing. Circular financing or or so it's now been called. Um and and this is something we've been writing about for the last couple of years. In essence what's happening is similarly uh Nvidia is uh financing or is fueling uh a new paradigm let's call that AI artificial intelligence and so in order to do that you need funding so what Nvidia is doing is it is investing in startups andor mature organizations that are um that need to purchase its chips to build out their infrastructure and create the large language model simulation uh systems. And so in order to do that, Nvidia invests, let's say, $1 or one, let's call it, you know, 10 million and then they'll get a number of other investors to invest another 90 million and you now have a hund00 million investment in this startup organization. Let's call it Open AI. OpenAI then turns around and purchases chips from Nvidia. So that becomes circular. I give you one, you turn around and buy products from me. Why is this interesting from an e that's that's the economic reality of what's happening? But on from a financial statement perspective, what's going on? Nvidia takes the 10 million, it turns that cash into a long-term investment. Okay. Now this investment could turn out to be fantastic because that company as you mentioned let's say take the case of open AI it went from 20 billion to 400 billion so on Nvidia's finan on Nvidia's investment financial statements they're able to book that increase in valuation so they're their shareholders are benefiting from Nvidia making this seemingly great investment and this continues but from now the financial statements they have this investment They record revenue. The products get sold. Once they're sold to an to a business in which Nvidia is invested in, are is that occurring at as an armlength transaction? Is that occurring at market rates? Do you think that those companies that Nvidia's invested in? Do they care what they're paying for the product when they stand to lose perhaps some more investment from them? And so they take that that uh in that investment. Okay. So now it's sitting on on Nvidia's balance sheet and there's no cost associated with that investment. This is different than what Nortell was doing. Nortell would in essence sell product and offer you time to pay for it. Nvidia is not doing that. They're they're they're making an investment in your business and you're turning around and purchasing product. >> So let's talk about Microsoft. Now they have a legacy software business but they also are building out these data centers and uh they own 27% of open AI and so when you look at these companies Microsoft is a publicly traded company but where do you see the risk there with this scenario because it seems to me when you look at open AI it has revenues of $10 billion but it's got financial commitments for $1.4 trillion. Okay. and they still have not provided a game plan on how they're going to provide for those commitments. Now, maybe their revenues are going to double or triple here in the coming years, but that remains to be seen. But once again, part of your job as a forensic accountant is identifying risk. Where do you see the risk in this scenario? >> The risk is that any business is only as strong as its customers. So if Nvidia, Microsoft is investing in startup organizations, um they're obviously banking on profits to occur later on. Now, because they're p they're those customer those those invest positions in also become their customers, they don't actually care if they don't make any money on their investment. But from the standpoint of the overall industry, what's what this what the weakness and the risk that I see is in order for open AI to continue to to to work andor perplexity, anthropic and so on, they need to eventually convince and investors that there's a view to a profit. And the longer that time frame extends, perhaps investors will become impatient. And there in lies the risk because if they become impatient then Nvidia's customers no longer buy product from them and then Nvidia's growth slows and then it circulates the other way. There's no investments. There's no market cap expansion. Margins contract >> and that's basically the same thing we saw happen back in the early 2000s when we had a pullback in the market. People just stopped spending. Nortell wasn't getting paid. No. Nobody was buying their product. >> The customers had to pay back that purchase that they made from Nortell. That means they needed to get capital from outside sources. If the moment that that capital dries up, the whole circle dries up. It always rests ultimately on liquidity. Liquidity is like oxygen, right? You don't know how much you need it until you don't have it. And right now Nvidia is fueling that with liquid its own liquidity as is Microsoft and benefiting as a result. But if that dries up from other sources, that's going to be a real problem. >> So you raised a very interesting point there, the importance of liquidity. And another hyperscaler that's coming under pressure right now is Oracle. It's down 40% from its top that we just saw here in the last couple of months. Y >> and one of the issues is these commitments it has with open AI. And one of the things that's come out here in the last few months is that the market is getting concerned about being able to monetize AI. And the hope was that when they build out these data centers, they're going to monetize AI with by 2027 2028. Now there's speculation it might not happen until 2030, which really impacts the whole profitability of these companies like Oracle. And the other interesting thing about Oracle is that their credit default swaps have really blown out here in the last couple of months. 125 to 130 uh highest it's been since 2009. But maybe you can just speak to that because once again it goes back to what you were saying. Everything is working until it doesn't and all it takes is one little crack in the foundation and income's falling down. you know, investors when they're investing in uh growth companies, they're buying dreams. And those dreams are based on stories. And if that story is look, we're going to be profitable eventually, we're going to change the world, and as long as that is believed, then everything's fine. It's when not when the providers of that story or the the organizations and the executives that underpin that story when they change their story as you said or um with open AAI and and Oracle having to change and extend out that uh view to a profit as I mentioned right I said earlier I said it depends on if if investors lose patience well if investors lose patience then we start having a problem and you got to following a bit the tea leaves, right? Everything matters. If we go back a few quarters, no one was asking about inventory turnover, AR turnover, changes in depreciation. These are things we've been talking about for the last couple years related to the players in this industry. And yet, no one wanted to read about it. No one called us. No one cared because the market was going higher. and price, okay, creates uh narrative. If the price is going up, I'll just justify by saying I'm buying a dream. The moment the price starts getting choppy, now people start asking questions. Wait a minute, when's that when's that earnings coming? When's that cash flow coming? What's going on with that inventory? What's happening with your margins? If that start the narratives start changing into that, then the story is changed. And when the story changes, then the multiple may have to be impacted. >> And you touched on depreciation, and this is another concern we've uh seen in the last few months. Michael Bur came out with a report and he's saying the depreciation associated with these chips. Uh it's too long. And he's basically saying the life of these chips because of technology is changing so fast that the life of these chips is much smaller. Yeah. Maybe you can just speak to that and how that's going to impact. >> I I think this is a this is an area of uh significant debate that only time will tell and there's a number couple of different factors that I would say. So yes, you know, Nvidia wants you to believe that they're short because they want their customers and the market investors to believe that you need to have the next best thing. But if you travel across the pond to the other side and look at what's happening in China and I recently visited um they created deepseek which uses a different algorithm to do its uh um analyses and as a result uses less power and works on old older dated chips from Nvidia. And I I haven't tested the speed, so maybe it's a little bit slower than than Chad GPT and and Perplexity and so on. But in when we're talking about something that answers you in seconds, if it takes a few seconds longer at this point, I'm not so sure that people are too worried. Um because there's going to be different uses for AI. Some that require instant answer and some that may not. So I think there's still debate. uh there is definitely the question that they do get burned out especially over high use that they've shown that they they lose their efficacy but at the same time I think there are ways that will extend out their life um and that's not a good story for Nvidia because if no one needs to invest in the new black well to get to the next who knows what then where does the next growth come from >> and if you have one concern from an accounting point of view about AI and all the companies associated with it, what would it be? >> This is not a question of uh you know Nvidia over capitalizing costs or or recognizing revenue too soon. I I I don't know. It's not the it's it's my concern is these are in effect related party transactions because uh Nvidia is is is selling to organizations and we've identified 80 of them uh that uh Nvidia is selling to and so are those happening at market rates and um if they are fine but if they are not uh then and and those organizations cannot get external financing um then Nvidia may have to sell to other organizations at different prices and that could affect its its margins. We don't know you you with with accounting or or when you're looking at financial results you end up with questions. You end up with looking at trends and situations that don't m don't match well with the economic reality. I hope that's coming through in what I'm talking about. And so then you continue to ask the question why. Remember I said what's the what does a forensic accountant do? It continually asks why. I'm a I'm actually a pain in the ass. When people tell me things, I always ask why because nothing happens without a reason. It's important for Nvidia and all the other major players to continue to invest in their customers because they need the AI to fuel the liquidity. It also gives a once you have the the the you can say that Nvidia or Microsoft or some other large mega company uh MAG7 company is invested in you well you use that as a badge of honor don't you want to invest and so again when the narrative changes that's what makes me nervous it's not something that I can pinpoint specifically this account on on Nvidia's financial statement makes me concerned >> so we started this conversation by discussing Nortell which was a Canadian darling back in the early 2000s. >> I want to ask you about the Canadian Telos BCE, Telus, Rogers and these companies peaked in 2022. Y >> and since that time both BCE and Telus are down 50% give or take. Yeah. >> What is happening within this sector? Because at one time this was the sector you would want to own for stability and growth, dividend yield. >> Yeah. In many ways they're a utility. They make an investment in infrastructure like a pipeline in or a or a power generation. In this case, they put in uh you know uh cellular nodes and they put in uh uh you know infrastructure cabling and so on and then they sell use time on that. And um this is actually something that we warned about. I mean we were first to warn in 2023 that there were cash flow problems at BCE. What I think has happened is uh the whole industry has in essence done it to itself. So you have a government that needs revenues and you have seemingly an insatiable demand from customers for internet capacity and mobility on both mobile and and on landline, right? And so what do you have? Well, we have 3G. We got to go to 4G. We got to go to 5G. Well, who sells that? Well, the government. government sells the space on the f in the um on the networks basically the the the data the waves and so the government there's no price for this it said hey I have this much of airwave available in 5G network and so the three major players beat themselves up to bid for that and paid a fee to the government and then the government says to them oh by the way now you paid for this, but I'm watching how much you're charging customers. Okay, you're going to have you're charging too much to customers. You're going to have This is not good. We're watching, right? And there's all this pressure on them not to increase prices, but it doesn't make sense. When they let Enbridge put in a pipeline or they get they let um Hydro1 build a a power generation facility, they allow them a a rate of return commensurate with their cost of capital. Why is it that they don't do the same for Telos? And so because they're under pressure, can't charge what is perhaps a return that they should charge to make money because Canada, you know, they say, "Well, Canada pays too much in mobility." But, you know, when I'm up in North Bay and I go to Sudbury and I go to Timmans and I turn on my phone and it works and it's fast, that costs money because there's very few people there. And you know, it's easy to put up cellular satellite sites here in in dense populous areas, but Canada's spread out so wide that that's where the cost comes in. Anyway, maybe I'm sounding like I'm promoting their businesses, but it's it's never a simple answer. And so that's that's how I think we got into the problem. Um, for the companies themselves, they overspent. Okay? And it's a it's a business that's always under pressure because we got three main players fighting against themselves largely on price. And so uh we saw that their cash flows were eroding. Everyone was buying them comparing them to the companies I just mentioned Hydro1 and Embridge for example. Well, if they're pay out paying out this level of yield and they're increasing their dividend each year, if I want to be considered as a comparable on the dividend funds and for everybody looking for dividends, I also have to do the same. And so, they were doing the same. But then it got to a point where they could no longer mask the decline in their free cash and they ended up were pay paying too much in dividend. And here we are. Rogers different because they didn't uh um sort of build out as much. They they made the big acquisition of Shaw and so that took added debt for them and they didn't increase their dividend as much. Their dividend yield always stayed lower. So I think that's what positioned Rogers in a better better from a cash flow perspective. Um it's been the one that we we like best. Uh and as well Quebecor, they bought up the the freedom assets, the mobility assets from from Shaw. Very good move for them. They took on that which generated significant cash flows, opened up a whole market to them. And there you go. >> The dividend yield on both BC and Telus is still relatively high. I believe it's eight or nine% and that's always a good reflection of that there's still trouble. >> You know, I I I have to make a plug here. I I I wish investors would stop aggressively, I'm saying this, using payout ratios tied to earnings. Earnings is made up. It's an it's an it's a it's a subjective number. Cash flow also has its subjectivities, but at least I can it it starts with and moves around actual cash flow. Dividends have to be paid out of cash flow. So the dividend when you calculate payout ratios what we do while it's amusing to look at the payout ratio r related to earnings we look at it tied to cash flows and that is where you'd see that the c payout ratio okay is too high more than 100% of the free cash they're generating and then you have to normalize the free cash because there's no textbook way of calculating free cash flow so each BCE Rogers and Telus calculate it differently. So we normalize those in order to come to a number such that we can say which one is more at risk than the other. >> The other thing I remember about BC in 2025 they closed on the Maple Leaf Sports deal. Uh they sold 37.5% to Rogers. Rogers now owns 75% of it. And in my mind like I'm thinking why the hell did you do that? Because if you want to own anything in this environment we live in now, it's sports content, right? And you look at Netflix, one of the greatest media companies in the last 10 years, what are they doing? They're going all in on live sports. They're paying $500 million a year for wrestling. They just paid $150 million for two NFL games on Christmas Day. And yet BCE just sold their stake in Maple Leaf Sports. What are your thoughts on that? You know, that's why debt kills. When when individuals and organizations overlever themselves, they can no longer make business the best business decisions. They have to make decisions to solve for a problem. And that's when uh there's opportunity obviously always for the buyer, but that's when you that that's why companies that are highly levered are higher risk. Anthony, I want to get your views on the Canadian economy and how it relates to the real estate market, especially here in the greater Toronto area. And I guess you could characterize the Canadian economy is weak. And the unemployment rate across the country, it's around 7%. In the province of Ontario, it's 8%. In the city of Toronto, it's 9%. And this trend has been going up for many months, many quarters now. And what does all of this mean for the real estate market across the country and more specifically just in the Toronto area? >> Let's break that down a little bit. I want to start with your comment that the economy is weak. Well, GDP is still growing uh >> moderately >> moderately but growing. Um the retail sales are weakening but still holding up. Um yes, we have seen an increase in unemployment which typically is a sign of weakness. But the problem with uh any metric is it's just a ratio. We've spent an inordinate amount of time focused on this in particular and um our analyst Shallab Guard's done a great job gone back to and this is tied to the real estate conference which we've been having for the last 13 years. We actually publish on a quarterly basis just a report specifically on real estate. So, we went back to 81 because that's when uh the great the greater Vancouver area uh rolled over and we went back to 91 because that's when the GTA rolled over. And we looked at what happened. Remember, we're always in the question asking why. Everybody thinks it was unemployment. Well, unemployment was already rising prices and prices continued to uh increase during those time periods. What happened ultimately was it was not until there were job losses. Job loss means Joe has a s has a job on in January in February he doesn't have a job. March he has no cash flow. Now we got a problem. Unemployment is a ratio of the total number of humans that are looking for a job over the total number of humans of age and ability to have a job. If we had under the previous guy that was sitting in Ottawa increasing our our num our numbers of immigration, well, we brought in many individuals and the number of individuals that came in was more than the jobs that were being created. So, we had unemployment, but those individuals didn't have a mortgage because they just showed up lately. Really hard for them to get a loan. They didn't have significant consumer debt because they didn't have the cred the credit quality to get it. And so, while the number on the surface goes, we're flashing, we got a problem with unemployment, when you look underneath, there aren't yet job losses. Now, they've just started. There were some job losses in July and August. And my concern, which ties to everything you've just said, is that real estate is now getting into a headwind. And that headwind is there's no new housing starts. When a building is finished, if there's no new building to go up, what do you do with the plumber, the electrician, the drywall guy, the form guy, the skipper who was on the job? They go from doing something to not. You can't build half a building. You can't do half a window. You'd build a whole window or no window. And so we have no new projects coming up coming some part in the middle of 26 and onward. That means there's going to be a huge headwind if we don't get infrastructure spending or something else for these individuals that are in the construction industry to do. And my fear is we have like 10% of our economy, 10% of our employment in construction related activities and that only includes those that report their income. Now I know this is not something you, your family, anybody in your area would do, but there are parts of this country that when they hire individuals and organizations to do their work in construction, don't ask for a bill. Those guys aren't being reported in that income number. And I hear stories that that number could be very significant. If that slows, you see a further erosion in the real economy and that's when I think there'll be a problem. So for now, we haven't seen any real problems. Mortgages have been coming up for renewal at higher rates. How is it possible people have been able to make it work? They still have a job. >> Very good points. And uh just to provide some more context, in 2025 over 1 million mortgages came up for renewal. And in 2026, I believe the number is 1.2 million. All right. So the Bank of Canada has been very aggressive in trying to get these rates down so people can finance these mortgages. >> And I'm sure they're very concerned about it because, you know, if your mortgage payment is $2,000 a month and then it goes to $4,000 a month, you got an issue. And to your point, if you lose your job, then you got a serious problem. And going back to unemployment, we still have this trade war going on with the US or >> it's not resolved. Let's put it that way. >> And we're seeing a lot of negative impacts. >> Stalantis just shut down a plant in Bmpton, 3,000 jobs gone. Then you look at the suppliers that feed that one plant, there's probably 3,000 more. General Motors closed down another plant where they were making that Bright Drop EV uh delivery van. That's another thousand jobs gone. Al Steel just laid off a thousand people, one-third of their workforce. So there's this big trickle down effect. And so when you look out into 2026 or later in 2026, where do you see the unemployment number going? And how does that impact the real estate market? >> Well, I I think the unemployment rate will rise and it'll rise from job losses. which is the danger which has the trickle down as you described to the overall economy. And the other interesting part about our current economic situation and this comes up in the US as well is it's a K-shaped economy. So think of the shape of a K. So those with assets since since uh COVID have done phenomenally well because the stock market's up three-fold. Uh their their house went up in value. They're doing great. Those that are living on paycheck to paycheck, I call that Joe six-pack. They're suffering because inflation is the death smell to an economy because what it and it affects everyone whether you're rich or you're poor. You show up and you see that the h that the the the stake used to be 20 10 bucks 20 bucks and it's 40 bucks. You say whoa even if it's not a big material month amount to you say whoa maybe I'll buy this small cheaper cut. And so what's happening now is if those on the bottom part of the K and and tied to the inflation, what happens is, you know, inflation they say, "Well, inflation is down. We're at 2%, Anthony. Now, isn't everything fine?" Well, it's a rate of change. So if year one we went up 8% and the next year we went up 6%. And now we're down at 2%. That feels good. But that forgets the point that we were 8 plus 6 plus 2 plus the amount the accumulation of that we're probably close to 20% increase in price to that human from 3 years ago. And so you see this we're at the latter part of the stage of the shift of the economy. It just came up in Air Transact. Why do you think that the the the pilots were willing to not work and they wanted and eventually got like a 60ome percent increase in salary? Because that's what they needed. They hadn't had a a a wage increase in years because they were a onesie, twozie, threes percent because that's what the inflation rate was at. That seemed to make sense. Now all of a sudden it no longer makes sense. And this I think is going to be the strain. the fact there's there's there's this strain between the worker and the company. The company willing to shut down all its airlines, its it aircraft and the and the and the workers willing to not work in order to uh because they're both at an impass. Now we're we're we're at that point where impasses are happening. That's when the real cracks come. I think we both agree the economy is weakening and we're seeing these unemployment numbers tick higher. But what's really surprising to me is when you look at the TSX, it's up 30% on the year. Just an amazing year. S&P still up 15%. So massive outperformance in Toronto or in Canada over the US, but uh a large component of the TSX are financials. I believe it's 30 to 35%. And there's been no weakness in these Canadian banks at all. Does that surprise you? Um no uh first answer and and secondly I think it's really important you know that people uh need to stop looking at the index be it S&P TSX uh the NASDAQ and go well if the Dow or the Dow the Dow's up 3% everything must be fine in the economy everybody must be doing fine or the S&P's up and therefore or everything's fine or the S&P's down and therefore everything must be bad. It's just a metric. 500 companies and that's how their valuation changed that day, that year, that month. It may have absolutely no uh link coincident or otherwise with what's going on in the underlying economy. the market is looking forward and it contains companies that may not be a reflection of the over outlying economy, right? Like real estate as a percentage of the Canadian index is virtually zero. We don't have any public real estate companies. So we're having a struggle in real estate in in real estate, but there's nothing to be affected in the index. Well, that's because there's no real estate companies. Do we have REITs that own real estate? Okay. Some have done decently well, some have not, but we don't have like constru construction companies uh that that build things on our index. Okay. And so that that's what I think creates this this uh dichotomy that you need to that investors need to be careful of. >> And so when you look at the financial sector overall, you're not concerned about the Canadian banks? Well, I listen, I'm I'm I'm concerned because their their valuations are at that the more uh they're elevated relative to historical averages. Um but uh I I'm I would be I'm getting I get more concerned as uh job losses occur. And I'm not we're not concerned because of mortgages per se, banks. Um you know, I I'd be I'm much more concerned about private markets. I'm very concerned about the private mix. I'm concerned about all private credit. I think that's where the the biggest risk and bubble lies. Um, but where I look at uh for for the banks themselves, it they'll lose money if they if it if we have problems in credit cards in unsecured lines of credit, those banks that have outsized exposure to those sectors of the consumer or those type of loans to the consumer are going to be more affected than those that don't. >> So overall, when you look at the Canadian economy, we we talked about the Telos. Do you still have some concerns there associated with a couple of the big players BC Intel? >> Oh, I would say I would say there could be there could be opportunity at some point like we're really getting washed out right we're we're down as you said significantly now it's going to be is is their business going to stabilize? I think investors once they see their business stabilizing, they'll be there'll be some turn uh because things have really gotten beat up here. And I think the government is more interested in other things other than your credit your uh mobility bill. They're going to be having to worry about unemployment such and so they're not going to be breathing down the big three's neck saying, "Hey, you better cut costs." >> Are there any other sectors within the Canadian economy that concerns you where you see a risk to investors? No, I think you you you touched on on on banks. I think the private space uh any any private equity, private credit, those are those are space that I would be concerned about. Um I would say uh consumer uh discretionary uh is one as well. We don't have a lot of that in Canada. Uh but those spaces are ones where we're going to see an issue because uh if we do end up seeing the job losses materialize uh then people just spend less you feel and then you have this the wealth effect in reverse even if you have assets and you have money you just think twice it just happens and so that I think is what often gets missed when economists look at numbers they just say well everything's fine or everything's not fine well you need to go on the And that's really something we do a lot of is actually go and shop, actually go and visit what a condo site looks like and what's happening. That's where we get some of our best information. >> Okay. So, you do a lot of work on condos within the greater Toronto area. What are your thoughts there? Because everything I read, it's a total collapse. >> Yeah, it's it's uh it's really bad. Um the the problem is and it's a slowm moving uh train wreck because you have uh many uh properties that have been purchased at high prices. They're diff they can't get the financing to close. they've uh overextended themselves many now there's too much supply and those can't they're not renting because at the price that the that the landlord wants because in order to make the numbers work they need a higher rent but but that is not what the what the what the uh tenant is willing to pay. So you have this impass among renters. You have the impass among buyers because the buyers think that the price should be lower. The sellers still believe looking backwards that the price is worth more. So you have this stage which I call denial. And so we're in the still the stage of denial. And there's five, it's like grief. There's five stages. The first stage is denial, right? Then you have anger. You call your agent. You get mad at your kids. your mortgage broker, your banker, everybody did something wrong, that's why it didn't sell, right? Then you end up getting depressed, right? And you you end up uh ultimately then calling back the guy that gave you the low bid and he tells you, "Look, I told you I was going to pay you a million. Now I'm willing to pay 800." And then the worst case, you fall into acceptance at the end and you end up buckling. And I my advice to people is, you know, if you don't need the property and it's not generating cash and it wasn't generating cash in the past, let's just get rid of it. Don't wait. If you if the cash doesn't matter to you, you can hold through the through the end. The the mortgage payments are not a big deal for you. Do the worst case scenario. It's not rented for a year. Can you still hold on? Yes. Okay. If the good location, then let's not worry about it. But if the location is not good, then start thinking about uh making different decisions. >> Well, Anthony, this has been a great discussion. I want to thank you very much for spending time with us today and sharing your thoughts on forensic accounting and the importance of it to investors. If somebody would like to learn more about you and your firm, Veritas, and the services that it offers, where can they go? >> Veritascp.com. That's the base. We're an independent equity research firm um covering mainly Canadian companies but we also do some some uh US companies and uh then we launched an asset management firm called Veritas asset management. We take those ideas and package them into portfolios. And then we launched a phil philanthropic uh platform where we offer donor advised funds and we do research on charities so that when donors make donations, they don't just give with their heart but they give with their with their mind as well and are armed with research. >> And Anthony, you also have a podcast. >> Yes. Who wants to listen to it? >> Yeah, our my podcast is called the factfinders. We started that uh in uh March, April of 2020 and it was really uh again my uh search for why uh the to go on the ground talk to industry professionals about what's going on in the economy. Uh and so our mission here is to help investors make better investment decisions. So we'll have different experts uh on the on the on our platform and uh you know try to offer that to our investors. >> Well, this has been a great chat once again. Thank you. >> Thanks for having me, Jimmy.