Four Key Macro Implications From AI Adoption

So, there are obviously a ton of people making AI content right now. There's endless blog posts, endless videos. I'm sure you're inundated with it. I'm sure you're tired of it. But, I'm going to make you one more. But, this video is going to be a little bit different. We're going to talk about the four key macro implications from AI adoption. So, this is something you probably haven't seen before. It's something we at the Macro Institute have put a lot of work into. We've been talking a lot about AI to our clients recently and this is just a small version of what we've been putting out. Uh so let's get into it.

So introduction, the AI craze. This is stuff you all know at this point. Artificial intelligence has exploded in popularity. According to a McKenzie study here on the left, up to 80% of companies today are using some sort of artificial intelligence. That might sound not that sort of staggering of a stat seeing as how much you've probably read about AI. However, if you look back in 2017, that number was just at 20%. So, there's a huge uptick in companies that are using artificial intelligence. And it's not just companies. A huge portion of adults are now using uh artificial intelligence for tons of different purposes. uh we hear so Google searches for AI or artificial intelligence obviously way up none of this is surprising I'm sure a lot of you guys use AI in some way in your day-to-day life whether personally or professionally and so this has required a large buildout so companies are pouring tons of dollars into AI infrastructure this is something that's been talked about a lot right now um their sort of $500 billion Stargate project Nvidia is investing a ton of money money meta Google, Microsoft, Amazon, you name it. Look at the chart. Up to $90 billion of just those four big companies combined. Um, this is clearly a priority, right? Again, nothing's here surprising.

But now, let's get into the list. So, this is the four key macro implications from AI adoption. Number one, AI has had an impact already on inflation trends, and that's only going to grow in the future. Number two, AI right now is not quite ready for prime time when it comes to financial analysis and research. We have some cool charts to show that. Three, it is causing employment anxiety. We'll have some key charts there that show on one hand, it might not be as bad as you think, but on the other, AI is probably coming for some of our jobs. Then number four, AI's influence on fiscal and monetary policy. Uh that one is a very macroy topic that we will get into at the end. And then also if you stick around I have a special offer. The macro institute is launching a brand new product line a macro investing boot camp that those who are here at this video uh we have a pre-launch offer uh that you'll find out about.

Okay. So number one AI's impact on inflation trends. And as you'll see this sort of goes both ways. So right now a lot of the inflationary forces that we're seeing are due from resource constraints and a lot of that has to do from electricity. There's a massive energy requirement from data centers. So if you look at the chart on the left that is electricity consumption uh and it's based on percent of US total if you look in the chart. And so if you look at the future scenario range um from about 2023 to 2028, this is from a study uh that was done a few years ago potentially in doubledigit percentage of total energy consumption could come from data centers alone. We've already seen the price of electricity spike as you can see here on the right the small chart within the chart we have utilities. This has lead it feeds through to consumer inflation. So consumers are spending more for electricity and other types of energy sources as they are now competing with these data centers. So AI can be an inflationary force.

On the flip side though, there are some economic mechanics of AI that could lead to deflation or disinflation and a lot of that comes from labor markets which again we'll touch on for number two as well. Here we show the inverse relationship of labor costs and productivity. So as you can see labor productivity is inverted here on the chart. So when productivity goes up so as people become more productive unit labor costs actually go down. So you need less people as people become more productive. So you are paying less and less for uh labor. That's what we see here on the right as well. So when labor costs fall you see CPI services fall as well. that is sort of a substitute for wages. So you might see increases in energy costs as well as downward pressure on wages. Now when you think about the consumer spending more on energy, potentially making less money, uh that can be sort of a scary backdrop.

Number two, AI not quite ready for prime time. So this comes also from the McKenzie study here on the left. AI risks organi organizations are working to mitigate. Number one by far was inaccuracy. I shouldn't say by far cyber security was also close but the top one is inaccuracy. It's considered the biggest problem right now with AI. Yet if you look on the right from that same study only about onethird of users are actually checking their entire output. So that would be like us. We write our weekly reports. we, you know, go to chat GBT, give the topics we wanted to write about, it spits out a report and we just hit send. We do none of that. Uh we actually don't use chat GBT at all for our reports. Um, but it's a little staggering that you know it it's acknowledged at these companies that inaccuracy is the biggest problem and not a lot of people are actually checking the entire output of AI. Um, so you know, from a financial analyst standpoint, from macro analysis, there's a little bit to go before AI is, uh, taking things over, I should say.

And so we actually tested this a little bit. Uh, so we went to OpenAI, Chad GBT, and we had them create a leading economic index for us, right? So we spend a lot of time modeling things. We create different um um investment strategies etc. Um this is what we do for our jobs. So you know if we could plug in and have Chad GBT make a better leading economic index for us um I would have been a little nervous but this is what it gave us. It gave us new business applications, consumer confidence, manufacturing new orders, housing starts, borrowing rates, job vacancies and stock market momentum. So, some stuff we're familiar with, some stuff we use uh in different LEI indexes, um some stuff the conference board uses, and then some other stuff that, you know, you wouldn't traditionally consider to putting into an LEI basket. And then on the right here, we didn't touch anything. We made no adjustments. That is the index it created. As you can see, it's not quite usable, right? This would require some additional effort from us if we wanted to use this index that chat GBD created. So it's can do a lot of things. It's getting close. But when it comes to highlevel macro analysis, macro research again as we say here, not quite ready for prime time.

Before we get on to number three, I did mention a special offer and we have launched or we are about to launch, I should say, a macro investing boot camp. So as some of you may know our flagship program is the macro specialist designation. It is a full designation program. Um I consider it sort of the the CFA is bottomup investment analysis. I would consider this a top-down version of that. Um so it's applied macro analysis in a professional designation. I have my CFA and I also have my MSD designation. So, one thing though, we know professional designations are a big commitment in terms of cost and time, right? When I was getting my CFA, I knew that was going to be three maybe four years of my time, thousands of dollars. It was a big commitment I had to think a lot about and I decided to go through with it. However, we think for our designation, we should have a introductory product or an introductory offer to see if this is the right fit for you. And so, that's why we launched this macro investing boot camp. It is 15 hours online self-paced boot camp that is a bite-sized intro to the mat macro specialist designation program. And here's the special offer. So, when we fully launch on October 20th, it is going to be $495. Anyone who pre-orders, so you guys are hearing this offer up front by watching this video, can save $100 or close to 20% off the price. If you want to learn more and if you want to pre-order to save that $100, I'm going to put a link in the description of this video that will give you all of the information you need. Last thing I want to say, if you do decide to go on to the full designation program, the 3.95 if you take advantage of this offer will be used to count towards your uh fee for doing the designation program. So, you will receive a discount from the designation program if you complete the boot camp based on how much you paid for the boot camp. All right, check it out. It's great. Packed with tons of information, 15 hours online, self-paced. You know, you could do it over a week or two and you'll have a ton of investment knowledge, a ton of actionable things you can work into your investment process. Okay.

And we are on to number three and that is AI has created a lot of employment anxiety. So, it's not all bad. We'll start with some good news. Artificial intelligence will lead to some new jobs. So, here I'm going to read this quote. Uh this chart on the left is the number of job postings requiring requiring AI skills from LinkedIn's future work report. AI related job postings grew 38% between 2020 and 2024 making it one of the fastest growing categories globally. Then we look here on the right. Nearly 10% of all finance job listings require AI skills. Now obviously this is a very specific skill set that not everyone has but the point here is to say it will create some new opportunities. Anytime there is some new technology or new boom um there are downstream effects where there are new type of jobs that don't exist now that will exist in the future.

However the big fear of everyone uh is not incorrect. Artificial intelligence will lead to the deletion of some jobs. So from that same McKenzie report, this pie chart on the left, um some firms forecast a headcount reduction based on AI, um almost one-third, so 31% said decrease. Um again, if you look at this, you know, we do have 40% that are saying no change and almost 20% that are saying increase. So it's not all bad. However, there are definitely more firms that are considering cutting jobs than adding jobs due to AI. And then here in the chart on the right we have the share of tasks that can be performed by AI. This was a study done by Wharton. Um 27% of jobs they found can be done almost completely by AI. So obviously if you're using AI a ton to do your job. Maybe that's not a great thing. I know right now it's probably making your life easier. But any of these jobs that fall into that 27% category where 90 plus% of it can be done by AI. um that could be a problem area and so what happens just quick simple fallout of AIdriven job loss so as AI gets widely implemented seems like we are going to see more jobs lost than created here according to CBS for the first seven months of 2025 adoption of generative AI by employers accounted for more than 10,000 job cuts okay so what happens is people lose their jobs consumers spending slows and prices decline. So, this could feed back all the way to what we were talking about um for number one, which was the inflation component of AI.

And then last but not least, number four, this is probably the most actual macroy of them, but the impact of productivity to the real economy. So, there's a positive relationship between total factor productivity and US real GDP. That's a bit intuitive, right? as we become more productive as a society uh we would expect that our GDP would grow and then from that same study by Wharton we can see according to them the productivity gains from AI will be frontloaded so this is done by percentage points to the level of uh productivity you can see that 2% um sorry here I'll just read the quote cumulative productive projected growth contributions applied to that level of TFP will be around 1.5% higher by 2035, 3% higher by 2055 and 3.7% higher by 2075 relative to the no AI path. So it should add to productivity and should add to growth and well that is important. So you've heard some politicians now specifically after the passing of the big beautiful bill um with concerns that it's going to explode debt to GDP. There's been whispers that maybe we'll try to grow our way out of it. Um, not here to comment on the feasibility from that, but can AI potentially solve the fiscal crisis? Um, sort of we better hope so because depending on the this is a simulation done by the CBO on the low total factor productivity world versus the high total factor productivity world. Vastly different, right? between 20 sorry 200% jet debt to GDP ratio ratio versus a 100% debt to GDP ratio. A much healthier outcome under their high productivity simulation.

And that brings us to the conclusion. So is AI a miracle coming to fruition? Well, if you look at the chart on the left, the answer seems like it's going to be yes. But you need to be a little bit skeptical. So, as we can see, this chart shows the types of tasks that AI can complete. So, here's a quote. If this trend continues, AI models will be able to handle multi-week tasks by late 2028 with 50% reliability and multi-day tasks with close to 100% reliability. Two years after that, they'll be able to tackle half of multi-month projects. Now, that could be true. However, on the flip side, my wife and I just had a newborn and we went to a doctor's appointment after a month and she had grown an inch. So, at that trend, she would be growing 12 in a year. So, by the time she was 9 years old, she would be 9 ft tall. Okay? So, obviously, stuff like height that levels out. At some point this incredible growth trajectory will not stay this incredible. Right? The issue is we just don't know what that level is. So here they're projecting that by 2030 AI will be able to tackle half of multi-month projects. Is that going to be true? It's really difficult to tell right now. However, you know, does that mean five years after that they're going to be able to do hundred year projects? I doubt that's probably true. So, at some point we will get a leveling off in how productive AI actually is. The hard part is we just don't know when that exactly is.

Recently, a new achievement. I mentioned I have my CFA designation. We are big fans here. The CFA at the macro institute. AI passed CFA level three uh with flying colors. So if you look here on the graphic got close to an 80% uh open AI their 04 MIDI Gemini got close to 76% on mock CFA level 3 exams. So you know presumably a lot of people who watch our videos here are in the financial services industry. Do I think that do we think that they're coming for our jobs imminently? No. Could they be sometime in the near future? Potentially. So, maybe it's worth it to brush up on your AI skills uh and make yourself indispensable. Okay, I'm going to stop there before I get too doom and gloom. Um, thank you for watching this video. Again, we have a special offer in the description of this video. $100 off our macro boot camp that launches on October 20th. I really hope you check it out. Subscribe to the channel and stay tuned for more video content on the horizon. Have a great day.