Amazon was founded in 1994 and went public in 1997. By 1999, some 5 years after the company was started, only about 1% of total retail sales were being done online in the US. So you have to give it to Bezos, he saw what was coming and he got in early to help create it. This was not so obvious back in the mid 90s. The internet as a whole was still being viewed with skepticism, especially after the dot-com bubble.

Today, online shopping represents over 15% of total retail sales. (See above chart from Charlie Bilello.) The pandemic pop is over, but it looks like we've returned to a pretty clear trendline -- up and to the right. I guess the questions now are: When and where does this start to flatline? It doesn't seem likely that this goes to 100% in the foreseeable future, especially if you include grocery. But it's going to go a lot higher.

For myself, if I were to exclude food/grocery, I would say that the vast majority (80-90%) of my retail purchases are done online. Even if I'm in a physical store, I'll often pull out my phone to price compare. If it's cheaper on Amazon, I'll just order it there.

Here's another example.

This past summer when I was in Park City, I discovered the brand Vuori. I had heard of them before, but I had never actually seen or touched their clothes. It's great stuff. But instead of the store convincing me to buy something, it convinced me that I like the brand and that I should probably shop on their website at some point in the near future. And that's exactly what I ended up doing. (Sorry Lululemon. You're still my favorite.)

All of this is perhaps obvious in a world where 15% of total retail sales are happening online. But I would imagine that the retail landscape and our cities will look very different when this number goes even higher. Our cities were different at 1% compared to today at 15%; so imagine what 50% or 80% might be like.

Back in 2006, online shopping in the UK represented about 3% of total retail sales. As of March 2020, right before the pandemic, this number had increased to about 22%. Online shopping then spiked during the pandemic, as we all know, but it has since normalized and the January 2022 figure was about 27.1%. But if you exclude grocery sales (because food is special), I believe the percentage of online sales is now in the range of 35-40% for the UK.

What is clear is that there is a longstanding shift towards more online shopping. It has just become so easy. However, UK retailers, as well as many others I'm sure, have been arguing that part of the reason for this systemic shift is an unfair tax advantage.

Physical retailers tend to be located in high-traffic areas and so they are naturally subjected to higher municipal taxes. Online retailers, on the other hand, get to locate their warehouses and distribution centers in lower cost locations and so aren't taxed at the same kind of rates.

Because of this perceived imbalance, the UK is now studying the pros and cons of implementing an online sales tax (OST) as a way "to help rebalance taxation of the retail sector." Tax online so offline seems more attractive.

But this is a tricky thing. Nobody wants to see our main streets and urban centers void of retail activity. That would be like walking through a financial district that had decided to bury all of its retail activity underground or something. But at the same time, does it make sense to shift taxes to what the market seems to like and want? Though I suppose it could be a lucrative tax since you generally want to target things with inelastic demand.

What are your thoughts?

More info on the OST consultation, here.

Benedict Evans asks some great questions in this recent post about ecommerce penetration. Instead of just looking at the product itself and/or the way in which we buy it (online versus offline, for example), he focuses on the logistics model that accompanies the transaction.

What can be parceled and shipped via Amazon? What can be delivered using a bicycle? What requires some sort of special delivery or collection method?

The point he is making is that different things need to happen for a new fridge to make it to your home, compared to say a Chipotle burrito. And these differences matter when it comes to how we should be thinking about ecommerce and the real estate in our cities.

Personally, I find it helpful to reframe the questions in this way.

Here's an excerpt from the post:

But if I buy online and then drive to the store to collect it, is that different to phoning and reserving it? We didn’t have a statistics category for ‘telephone ordering’. If I use an app to order pizza instead of phoning the restaurant, has that become ‘ecommerce’ or is it still pizza delivery? 30 years ago, if I drove to Walmart instead of walking to a neighbourhood store, or drove to Best Buy instead of going to a department store, we didn’t call that ‘car-based commerce’. So is this a tech question, or a retailing question, or an urbanism question?

For the full thing, click here.

Chart: Benedict Evans