[hirvi74]

> Inflation acts as a tax on savers and a subsidy for debtors.

Slightly tangental, but you piqued my interest. If it's clear the Fed can print more money, which causes inflation, which is a tax on savers. Then why do savers or anyone need to pay taxes? Could the Fed just not use inflation as a form of taxes in general? Instead of taking x% of people's income, the dollar could just print the money they need, and inflate the dollar's value to recoup their costs.

I'm not saying this is a good idea or anything. I am not knowledgable enough in economics to argue for/against this. It's just a food for thought kind of thing.

[eru]

> Could the Fed just not use inflation as a form of taxes in general?

In some sense, they already do that.

But it's more complicated than it first appears.

First, most of the money the Fed+government makes from issuing money comes from seigniorage, not from inflation.

Simplified a bit, seigniorage just means that the cash people hold in their wallets (and bank accounts) took the government pennies to print, but the people offered real goods and services in exchange to acquire it.

Second, higher inflation makes people hold less cash in real terms. A simple illustration: when Germany had hyperinflation in the early 1920s, some people might have carried cash by the wheelbarrow, but even a whole wheelbarrow would only be worth eg a few apples and potatoes.

In contemporary Germany with a stable currency, it's not too unusual to carry enough cash in a slim wallet to be able to afford wheelbarrows full of apples and potatoes.

So if the government+central bank want to maximize how much real benefit they get from printing money, they can't just crank up inflation. In fact, lower inflation is probably better, if you want to maximize this.

[mosburger]

I'm not even close to an expert on economics, either... but I think what you're describing sounds similar to what's called "Modern Monetary Theory" or MMT.

If I understand the very basic gist of MMT correctly, it's basically that the government "invents" currency simply by spending. If the government wants to finance, e.g., an enormous infrastructure project, all it has to do is will it into existence and the money will get printed - mostly because the full faith in credit of the United States Treasury will make sure the right people are paid.

If this is true, what then is the point of taxes? Taxes are a buffer to "sop up" excess cash in the economy to keep inflation under control.

Another central tenet of MMT is that the end goal of tuning these knobs should be 0% unemployment, because that's when your economy is producing maximally.

I'm sure I got a lot of that wrong, because (like I said) I'm not an expert, but what you described reminded me of the MMT Wikipedia rabbithole I ended up in a few years back.

[eru]

Yes, you described MMT about right.

Unfortunately, MMT is either a meaningless tautology that doesn't change anything about our understanding of the world. Or, it's wrong. Depending on what definition the MMT people use at any one point in time during a discussion.

[nostrademons]

Heh, you've asked an eternal question which I asked in my intro macroeconomics class, and which I suspect is being asked at the Presidential/Prime Minister and Central Bank levels in many developed countries now. I'll give you both the macroeconomic answer and the historical answer.

The macroeconomic answer is yes, assuming that you can control inflation and spend your money in a wise disciplined manner, you can do this. An inflation rate of 10% is basically a wealth tax on currency-denominated assets of 1/1.1 ~= 9%, collected implicitly. And it has many benefits. Administration cost is near-zero, the government just needs to print money and let price levels do the rest. As a wealth tax, it's progressive. It encourages consumer spending, and encourages people to hold productive assets rather than currency-denominated financial instruments. Aside from land & externality taxes, it's one of the best types of tax.

The historical reality is much nuanced, and negative. In reality, you never get uniform inflation across all prices; instead you get something called Cantillon Effects [1], where money pools at the places it's injected within the economy and at monopolies, and never makes it to ordinary consumers. This ruins a lot of the progressiveness of inflation: Google and Facebook shareholders will enjoy higher ad prices (which is happening now), but the ordinary service worker on the street still lacks the bargaining power to get a raise. And it's also easier to dodge than governments believe: instead of holding wealth in cash, wealthy individuals will simply hold stocks, cryptocurrency, art, NFTs, and other "floating" assets whose value rises along inflation (this is also happening now). Inflation also acts on a tax on the economic efficiency of businesses and consumers - instead of the government collecting the tax, businesses simply have to re-price all of their items, and consumers may need to find alternate sources if relative prices change.

The worst case, though, is that it's unclear whether it's possible to maintain steady, government-controlled inflation at appropriate tax levels (which would have to be about 50% to maintain government spending levels of about 35% of the economy). Looking at incidents of high & hyperinflation [2], it's very difficult to find time periods where inflation exceeded 20% but then did not go on to exceed hundreds of percent. Businessmen seem to have a binary approach to inflation - either they can think of a world of stable prices (where maybe they give 2-3% CoL increases each year), or they must grab all the money they can right now because all their costs are going up too and if they don't they'll go out of business. There are many historical examples where governments thought "Okay, we can tolerate just a few percent higher inflation to fund this one war, and the spoils will pay it back later" and instead they found themselves tipped into hyperinflation and unable to rein the economy back in.

[1] https://mattstoller.substack.com/p/the-cantillon-effect-why-...

[2] https://en.wikipedia.org/wiki/Hyperinflation#Examples_of_hig...

[eru]

The problem with inflation as a tax is not so much that you can't target 10% exactly. We could probably do that these days with level targeting.

No, the real problem is the deadweight cost:

The higher your inflation, the less cash people want to hold in real terms. But that amount of cash is what you collect your inflation tax on.

> An inflation rate of 10% is basically a wealth tax on currency-denominated assets of 1/1.1 ~= 9%, collected implicitly.

What makes you say so? Expected inflation would purely be a tax on cash. Currency-denominated assets like loans and bonds would just get a higher interest rate up front.

Your next paragraph describes exactly that 'dodging' of the inflation tax.

Btw, you can also simply dodge eg a USD inflation tax by holding Swiss Franks. No need for NFTs.

(Of course, if you have a capital gains tax levied on nominal gains, then inflation effectively increases the capital gains tax. The solution here is to charge capital gains taxes only on real gains. That's a real problem in the real world.)

> In reality, you never get uniform inflation across all prices; instead you get something called Cantillon Effects [1], where money pools at the places it's injected within the economy and at monopolies, and never makes it to ordinary consumers.

The Cantillon effect is controversial to say the least.

In order to work like Internet Austrians commonly describe the effect, people have to be idiots who act purely on historical data and do not form any expectations.

In the real world, when the Fed announces some future policy in advance (like a taper or a new round of QE), that announcement has effects on eg the stock market right away, even when the Fed hasn't added or removed any money yet.

In contrast, for the Cantillon effect to work as described, money would need to act sort-of like a liquid or gas and 'slosh around the economy' and get stuck in nooks and crannies.