Here is what I learned this week - Europe is in more trouble than I first thought.

Those that are close to me know I am harbouring some big concerns about the world economy. In many ways this year feels very 2007 to me, but in some ways it also feels different. However, the undercurrent is the same. While the much vaunted public indicators such as the stock market is at all-time highs, when you talk to regional main street, small businesses and the average punter they are talking a different story. Inflation and cost of living is now crippling an already slowing economy, people are becoming more frugal - spending less and saving more and jobs growth (ex-Government) is stalling. There is also another major undercurrent - how hard it is to run a business today. Not just because of the economic backdrop, but with the continual bloating of the regulatory burden being pushed onto businesses. Everything from hiring talent to getting approvals to build infrastructure.

This week I was in London (is that technically Europe anymore? :-)) for Capacity Media - a techoraco brand Capacity Europe. A gathering of all people telecoms, data centre and a bit of satellite. There's also some private equity around too. It's always a great event with a few thousand people attending. You get the chance to meet, buy, sell, talk and share project updates.

Coming into this conference I believed Europe is going to be heading into recession, deeper and sooner than most economies in the world. Many of the reasons are are you would expect including labour laws that make it an unattractive destination for "global" jobs, jobs which can be placed anywhere in the world. The could be skilled or unskilled labour. Regional unrest including war in Ukraine and the Middle East soaking resources and challenging supply lines. Add to that a dislocated energy transition that has forced higher costs onto business both in actual power costs, but also capital costs to comply with new regulations that have little genuine effect on global warming but making businesses more unproductive and disincentivised investment in the region for global companies.

But by far my biggest concern was the entry of China as a serious competitor in the EV car space. In recent years China has developed good looking cars that are between 70-85% cheaper than the luxury brands ICE and EV vehicles with little material difference in performance and quality. I believe this is going to have a massive impact on Europe's single largest export - cars. Mercedes, Porsche, Audi, BMW in Germany alone are already starting to feel the winter chill of the winds of change. Volkswagen has announced intentions to undertake plant closures in Germany, because German production plants simply cannot deliver a car at a cost that is competitive in price to China. Already production is slowing, inventory levels are increasing. Job losses, plant closures aren't far away as well as profitability of European car manufacturers. The scale of this cannot be mistaken. The EU is in trouble.

But what you also get is a sense of undercurrents and a temperature check on the industry and the economy in general.

How could it be worse that this?

Well let's go back to Capacity Europe. Not only are we buying, selling and updating projects, but we are talking, having side conversations, drinks and a few dinners. It's the side conversations where you get a greater view of what's going on. And here is my assessment.

Europe is going to miss out of the vast majority of the AI spoils.

You speak to industry people in Asia and the US and you will hear of an insatiable investment in assets for all things digital infrastructure. Data centres, massive hyperscale fibre builds and overbuilds, new subsea hypercables between countries and regions to create petabit fabrics across cities, countries and oceans.

The same thing is happening in Europe, but just not to the scale. Why?

It technically matters little where AI training and inference is conducted, so it will gravitate to the easiest, quickest and lowest cost option avaiable.

Europe is making it too hard, too expensive and too risky to invest. Speaking with industry leaders they are all explaining that they are focusing the vast majority of their spending in the US because that's where the hyperscalers are investing. Second most focused investment is Asia/India, third is Europe and then there is the rest of the world. It seems like a 50/20/15/15 split between US/Asia/EU/Rest of world.

The US is home to Microsoft , Google , Meta , Amazon Web Services (AWS) , NVIDIA , Tesla , SpaceX and OpenAI etc.. They are exceptional in their innovation and demonstrate how the US is the absolute centre of innovation. In the US these organisations have near unlimited access to capital and talent. They also have a reasonable level of regulatory certainty and access to lawmakers. There is [near] unlimited capital available and open to investment that's needed to continue to grow the underlying infrastructure exponentially including in new energy sources from private industry. The US alone will likely get half of the world AI investment and continue to lead for the next 20 years.

Asia (including India) has the eyeballs with over two thirds of the worlds population, a path to continued growth as a number of developing economies keep developing and upskilling their population, a much lower cost base and faster approval times. Building infrastructure is generally much easier, quicker and cheaper in many countries in Asia (sans Japan). Asia has it's challenges around geopolitics and some countries have varying issues around protectionism, corruption and varying level of infrastructure maturity. However, countries in Asia that are solving these issues are being rewarded as evidenced with the massive investment in data centres in this region.

Europe obviously it's still a massive part of the worlds economy and always will be. But it's hard and getting harder. Between future EU regulatory risk on AI data and models used for training and inference especially around retention, data privacy laws (which are well intended but a nightmare to manage), lack of confidence in future energy availability, already high energy costs for new data centres in key markets, excessive red tape frustrating infrastructure investment, hyperscalers are placing more of their "chips" elsewhere.

Here's hoping I'm wrong, but as they say, hope is not a strategy.