It’s time for another research update. In this letter, four things:
- Updates now online
- Talking about inflation
- Dueling ideologies
- My Linux build
Updates now online
After the last research update, a patron suggested that in addition to sending out an email, I make my updates available online. I think this is a good idea, so I’ve decided to post each newsletter to Economics from the Top Down. The link (in the update title) will take you to the online version.
The page is public, so you can share the link if you like. However, I won’t advertise the research update page anywhere on my site, as I want it to be for patrons.
Talking about inflation
As a scientist, I rarely write about ‘current affairs’. What concerns me most are long-term trends, not the latest news sensation. That said, last fall I decided to write about inflation, which, at the time, had just started to rear its head. I knew the topic would garner attention … and it did. ‘The Truth About Inflation’ is now my second most-viewed post ever.1
The goal of that piece was to convey some basic facts about inflation that rarely make the news. First, inflation is never uniform; it is always differential. Second, because inflation is differential, it always creates winners and losers. The winners tend to be big corporations, and the losers are … everyone else.
Intrigued by this post, researcher Josh Levent asked me if I was interested in participating in an Interintellect ‘salon’ about inflation. Initially, I was skeptical, because Interintellect’s business model is basically to paywall a zoom conversation. But after agreeing to record the discussion, I decided to go ahead. You can listen to the full salon below:
The conversation was wide ranging, but began with a brief presentation where I updated the charts from my blog post. The summary: inflation is still differential and still redistributing income.
Throughout my academic career, I’ve struggled with how to best characterize mainstream economics. Is it ‘merely’ bad science? Or is it something more insidious — an ideology cloaked in the veneer of science? I’ve concluded that economics is mostly the latter.
Driven by this view, in 2020 I decided to dissect the language spoken in economics textbooks to see what it would tell us about economics ideology. The results, discussed in ‘Deconstructing Econospeak’, showed that economics textbooks systematically neglect the language of power.
After writing that post, I decided to apply the same method to historical data. The goal was to use word frequency to chart the rise of economics ideology over time. It didn’t take me long to get some results. But for many months, I sat on the data because I was unsure what to make of it.
Backing up a bit, to track the ideology of economics, I looked at word frequency within the Google English corpus — a massive database compiled from Google’s cache of digitized books. In this corpus, I measure the word frequency of ‘economics jargon’ — words that are overused by economics textbooks relative to mainstream English. (See Deconstructing Econospeak for details.)
Plugging these jargon words into the Google English corpus, I found that over the last four centuries, economics jargon became more common. That was expected. What was unexpected is that after 1980, this trend appeared to reverse. Over the last four decades, the frequency of economics jargon has dramatically decreased.
Why is this unexpected? Well, the standard story is that the 1980s ushered in the ‘neoliberal’ era of capitalism — a period when all things free market became en vogue. You’d think that this neoliberal turn would coincide with an uptick in economics jargon. And yet the opposite seems to have happened. For a long time, these results sat in my file drawer because I didn’t know what to make of them. But in the last few months, I’ve revisited the data and convinced myself that the jargon reversal is both accurate and important.
I’m in the midst of writing up the results, so I’ll save most of the data for my finished post. (I hope to have that finished in the next week.) But in the mean time, here’s a sneak preview. Once I decided to treat economics as an ideology, I realized that the most important thing to do was to compare the rise of this ideology with the fall of the old worldview. And so I decided to contrast the rise of economics jargon with the (presumed) fall of feudal jargon. But where, you ask, do we find feudal jargon? We look in the main canon of European feudalism — the Christian Bible. (Bibles, actually. I’ve analyzed the language in two dozen modern translations of the Bible.)
Here’s where things get interesting. When I crunched the numbers on biblical jargon, I found that they mirrored the numbers for economics jargon. In other words, as economics jargon became more popular, biblical jargon fell out of favor. And guess what … the reverse is also true. When economics jargon began to decline after 1980, biblical jargon began to rise.
I find this trend to be ominous. One of my hopes is that we can replace economics ideology with a scientific understanding of how society works. Given that you read my blog, you probably think the same way. But let’s be honest; scientific thinking is difficult. The easier route is to accept simple answers decreed by an authoritative text. And it seems people are increasingly turning to the Bible (rather than economics textbooks) for their (wrong) answers.
For me, recent US Supreme Court rulings have driven home this biblical turn. And it makes me worry about the future. Could it be that what lies ahead is neo-feudalism?
My Linux build
On to some personal news. As you might know, I am a huge Linux geek. In the last few years, I’ve had it on my to-do list to build a custom Linux install — one that did exactly what I want (and no more). This spring, I finally had time to make that happen. If you’re interested, here are the details.
To build my system, I choose to use Arch Linux, a rolling release distribution that is famous for being customizable.
Backing up a bit, newcomers to Linux are often surprised to learn that there is no single ‘Linux OS’. Instead, there is a Linux ‘kernel’ on top of which lives a huge open-source ecosystem. A Linux build starts with the kernel, but from there, you have a lot of choices. You can install a complete desktop OS (the most popular of which is probably Ubuntu). Or you can use a more bare-bones approach and build your own system. Arch Linux takes the latter approach. When you pop in the installer, you get nothing but a command line prompt. From there, you can build your system the way you like it.
A decade ago, the prospect of building a system from the command line would have scared the hell out of me. But I now have enough experience programming that I found the process quite fun. Basically, I wrote a bunch of shell scripts to install all the software I want and configure it the way I like. Although this took some time, the advantage is that I now have a customized system that is easy to reproduce. (If you’re interested, my install scripts are here. And my ‘dotfiles’ — unix-speak for configuration files — live here.)
Here’s a rundown of the software I’m using.
Sway windows manager. Over the last few years, I’ve moved most of my workflow to the command line. When you do that, you don’t really need a full-fledged desktop environment. Instead, what you need is a simple program to manage your windows. Sway is what’s called a ‘tiling windows manager’ — the idea being that it tiles your open windows instead of having them float. If you have a mouse driven workflow, tiling is quite annoying. But with a keyboard driven workflow, tiling is wonderful. Sway is basically an updated version of the popular i3 windows manager, one that is built to use a more modern protocol for displaying windows.
Neovim. The vast majority of my computer time is spent either writing or coding. To do that, all you need is a text editor. On that front, I’ve recently been bit by the Vim bug. If you’re a programmer, Vim likely needs no introduction. It’s a ‘modal’ text editor that is designed to let you do everything without taking your hands off the keyboard. Vim is famous for being utterly opaque to new users. And I have to say, the first time I opened Vim, I was completely confused. But now I’m a Vim convert … so much so that when I type in a non-Vim environment, I have to consciously avoid using the Vim keybindings. Neovim is basically a fork of Vim that is a bit more modern.
Alacritty. Alacritty is a sleek terminal emulator that I like for two reasons. First, it looks good out of the box. (The interface is bare bones … no menu, no scrollbar. Just a command line prompt.) Second, Alacritty is really easy to configure. You can set it up the way you like using a YAML configuration file. (For example, I have it setup to use a dark theme for coding, but a light theme for writing prose.)
fzf. I no longer use a GUI file browser, which means I need a fast way to get to different directories on the command line. fzf is a fantastic utility for that. It’s a ‘fuzzy file searcher’, which means you can type the gist of what you’re looking for, and fzf will find it. For example, every morning, the first thing I do is head to the directory of the blog post that I’m working on. fzf takes me there directly … no clicking around, no fuss.
Firefox. After years of using Chrome, I switched back to Firefox last year in an effort to wean myself off Google. (I left Firefox a decade ago because it was far slower than Chrome. But since then, Mozilla has modernized the browser, and Google has turned Chrome into more of a data-harvesting machine than a browser.) What I’ve come to understand since switching back is that Firefox is basically the last browser stopping an internet monoculture.
What many people do not realize is that while there appear to be many different alternatives to Chrome (Edge, Vivaldi, Opera, Brave), this choice is mostly superficial. Under the hood, these browsers all use the Blink rendering engine — which is written and controlled by Google. (Safari uses WebKit, an engine that shares a common heritage with Blink.) Firefox’s gecko driver is the last rendering engine that is not controlled by Big Tech.
Why does this matter? Some history is pertinent. In the late 90s, Microsoft used its power to gain near monopoly on the browser market. That allowed it to ignore HTML standards and institute features its own way. This hubris then forced developers to hack their websites so that they worked on Internet Explorer (but not necessarily on other browsers).
There are signs that the same thing is happening today with Chrome’s rendering engine. Developers design their sites to be rendered on Chrome, regardless of whether this fits with HTML standards. The lesson is that if we want to fight Google’s monopoly power, we need Firefox to survive. (Here’s a good YouTube video making the case for Firefox’s survival.
Back to my Linux build. It took more time than anticipated, but I learned a lot along the way and the results are extremely satisfying. If you’re interested in how I use my tiling setup, here’s an example. The image below shows my layout for writing. On the left is Neovim where I edit the document in Markdown. In the middle is Firefox serving a document preview (in HTML). On on the right is a terminal for doing useful things. I use Pandoc to convert the markdown to HTML, and have a script bound to a hotkey that runs Pandoc and then refreshes Firefox to show the updated content.
The nice thing about the Linux world is that it allows you to have this kind of ridiculously customized setup, if that’s what you want. Or, if you want something more standard, you can have that too. (For many years, I used Linux Mint because it had a Windows XP type layout.)
If you’re interested in Linux and the wider open-source world, I recommend the podcast Linux Unplugged. It’s funny and informative.
Until next time
That’s it for this update. Thanks for supporting my work and have a great summer.