Libraries versus Amazon

This is Joseph, talking about something Mark Palko tweeted.

There is an article suggesting we replace libraries with Amazon:

Amazon have created their own online library that has made it easy for the masses to access both physical and digital copies of books. Amazon Books is a chain of bookstores that does what Amazon originally intended to do; replace the local bookstore. It improves on the bookstore model by adding online searches and coffee shops. Amazon Go basically combines a library with a Starbucks.

The justification seems to be at least partially about taxes:

Now, I have a few observations.

The bookstore/coffee shop model (perhaps we could call it Barnes and Noble) is a nice retail model that has had some modest success.  But it seems to be in decline more than growth these days and presumes that the marginal person in it is actually a paying customer.

I will note that Long Island must have unusually expensive library taxes, as reports across the US in response to this usually report taxes that are an order of magnitude lower.  That said, locating things in New York City and environs is often very expensive.  I am not sure we should set national policy on the NYC real estate market.

It is clear that bookstores would like libraries to go away.  For the same reasons that Uber is probably not a fan of public transit systems -- if you reduce competition that is going to increase profits on the margin.  Is it really the case that subsidizing Amazon is an urgent public priority?  I mean I like Amazon and use their services, but they don't seem to be struggling as a company.

Finally, taxes are complicated issue but there is no example I can come up with of a large society that functioned without them.  There are some exceptions on the margin.  Medieval Iceland, for example, is interesting but obviously used a lot of family mechanisms to cover the gap.  Or the Incas, who may not have levied taxes per se, but had what looks a lot like a centrally planned economy to compensate.  Now it is clear that taxes can be too high.  But it is also clear that no central government often works out poorly.

What I find interesting is how people always want to try these experiments in the United States, a large, diverse, multi-ethnic country with a large military organization.  You can make a low tax version of this work, but I am wondering if we should try and make this model work for a smaller country first.  Maybe Iceland could try regressing back to the middle ages for a couple of decades and we'll see how it all turns out?