Comment by joe_the_user | Hacker News Reader

Comment by joe_the_user | Hacker News Reader

The paper disputes the idea that people’s expectations for future inflation matter much for the level of inflation experienced today.

What I have been reading lately gives me the impression that different patterns of capital allocation have changed inflation dynamics.

There's excess demand for older-generation used by the auto industry. In text book economics, that would lead to higher prices and then more factories and the meeting of that demand at a higher level.

But what happens is chip makers, the only ones apparently capable of making even older generation chip factories, won't allocate capital for such factories because they view as on their way regardless of demand.

So essentially, with huge capital allocations, companies won't chase short-term demand surges because they don't want to be left with future excess capacity (which is more costly than any immediate payoff). Look at Nvidia actively fighting the demand of crypto miners for their products. This seems to be a barrier to any inflationary spiral (increased demand leads to shortfalls, not higher prices and production. Maybe prices too but increases not propagating).

So, it's not that the economy is incomprehensible but it's structure changes and so it's structure can't be captured by pure economic theory.

This has always been the case. Economists are great at explaining the economy in retrospect - oh if demand rises we'll see extra supply. Oh except the suppliers can predict peak demand and won't expand to meet unsustainable demand causing a shortfall in supply even under excess demand. Sure, it sounds obvious, but actually predicting it rather than observing it is what's valuable.

> So, it's not that the economy is incomprehensible but it's structure changes and so it's structure can't be captured by pure economic theory.

This reminded me of General Relativity and the problem of unifying it with Quantum Mechanics. QM is defined on a fixed background, while GR says that background is dynamic.

Seems we need the Quantum Gravity version of economic theory, which tells you how interactions happen in a dynamic background.

> So, it's not that the economy is incomprehensible but it's structure changes and so it's structure can't be captured by pure economic theory.

Pure economic theory is an oxymoron, other fields dealing with human behavior figured this out a long time ago, while economics desperately clung to the idea that it was actually math, rather than a social science.

(Otherwise I agree with you)

just a small nitpick in the overall correct train of thought - is that it this reduction in output still does drive up the inflation - as fewer cars are made, their price increases (and it has knock on effects of increasing prices to anything affected by cars - delivery, logistics, you name it)

Except if the increase in car prices is known to be temporary than it doesn’t really drive overall inflation. I for one just fixed a car I would otherwise have replaced, due to the used cars I would replace it with being in short supply.

Inflation is a general rise in prices, not a rise in the price of one or a few goods. In order for prices to rise generally, there has to be an increase in the money supply relative to the goods and services it represents.

If inflation is reduction in broad (not one single item) purchasing power of the monetary unit, then inflation will be felt when people lose confidence in the money or the organization backing the money. USA will feel significant inflation when the public loses confidence in the federal government. EU is much further down that path.

People never lost confidence in gold, and yet the 1849 gold rush caused inflation. Gold's buying power declined significantly.

That is a good example of a commodity money that experienced supply volatility and subsequent price volatility. It was inflation - more supply of the money.

Money is tricky to do in all times for all use cases. Bitcoin is making a run for store of value money, and unlike gold it is truly scarce.

Inflation is whatever you want to to be.

Usually inflation is defined using the prices of a basket of goods. This does not capture the inflation that people feel - most people are far more interested in the price of a home than the price of spaghetti, but spaghetti is the definition. One reason why spaghetti is used rather than house prices is that it is less subject to speculatory swings, or local variation - to be used as a basis for adjusting other prices inflation has to be defined in terms of something stable and uniformly priced. This may seem circular, and it is.

What we see is that increases in the money supply generally don't go into the prices of basic goods tracked by inflation. Instead they go into the prices of assets that are not tracked by inflation. Nobody is using spaghetti as an investment vehicle.

what you're both saying is true, so where you think you are correcting him, you're not.

the inflation of the 70's was caused by the increase in the price of a single good, oil, due to the Arab oil embargo.

Oil being a factor in the cost of many other products, there was a knock on effect causing a broad increase in prices across the board, general inflation.

There were also 2nd order macroeconomic effects caused by the govt's attempts to deal with the inflation and economic slowdown, but at the root, price inflation was necessary to reflect the realities of factor prices in the economy.

reducing the money supply would not have worked to curb that inflation (nor btw did it work in the Great Depression, though it did bring about some deflation. Deflation is why bitcoin is not going to ever work.)

> the inflation of the 70's was caused by the increase in the price of a single good, oil, due to the Arab oil embargo.

If the price of a single item goes up, and you buy it, you have less money to spend on other items, and so the prices of those other items goes down. Supply & demand.

In order for a general price increase, the supply of money must increase.

> If the price of a single item goes up, and you buy it, you have less money to spend on other items, and so the prices of those other items goes down.

Almost no one buys crude oil directly.

But it is a production input for a lot of things, so a shift in its supply curve shifts the supply curve of lots of other goods (including most consumer goods) in the same direction.

> In order for a general price increase, the supply of money must increase.

For demand pull inflation, supply × velocity × average propensity to consume needs to increase; for supply push inflation, though, nothing is required on the money (which is demand) side, you just need on average decrease in supply of consumer goods and services.

Again, if you have $10, and the price of X goes from $5 to $7, now you only have $3 to spend on other things, which reduces the price of those other things.

But if someone gives you $20, now you have $7 to spend on X and $13 to bid up the prices on other things.

Demand push and supply pull explanations do not account for this, and are inadequate (though still popular) theories.

> Again, if you have $10, and the price of X goes from $5 to $7, now you only have $3 to spend on other things, which reduces the price of those other things.

Only in the bizarre world where prices are set by demand alone, and quantity traded is held fixed by the Hand of God. Real supply curves aren't horizontal lines except in very bizarre circumstances.

In a real economy, a restriction in supply of a widely used input with no significant change in money supply, velocity, or relative spending vs. saving desire causes the price per unit of consumer goods to go up and the quantity traded to go down.

> Demand push and supply pull explanations do not account for this

No, they don't “account for” your fantasy that consumer goods behave such that a price change for one good has the results expected only if all other goods had perfectly inelastic supply, because perfectly inelastic supply isn't normal.

It's useless trying to deny supply&demand setting prices. The law holds whether it's a personal fantasy of mine or not.

If you can explain how increasing the price of oil causes money to appear in my pocket to pay for it, I'm interested.

P.S. Nowhere did I say that supply was inelastic or prices are set by demand alone. I said "supply and demand".

> It's useless trying to deny supply&demand setting prices

And yet you keep positing a model which entirely ignores the effect of supply, by pretending that quantity of consumer goods traded is fixed constant and prices are set only by demand (i.e., assuming no changes in preference, supply of money among buyers), so that with a given supply of money being spent, increase in price of one good must be offset by decreases in other goods rather than decreases in aggregate quantity traded.

> The law holds whether it's a personal fantasy of mine or not.

Yes, despite their basis is known-false rational choice theory, the so-called laws (which aren't at all laws) of supply and demand are a reasonable approximation of real behavior of markets in aggregate.

But they don't predict the behavior you describe unless you assume perfect supply inelasticity of all goods other than the one whose price is assumed to increase.

> If you can explain how increasing the price of oil causes money to appear in my pocket to pay for it

It doesn't, decreasing the quantiry of oil traded at any price (increasing the price at any quantity traded) also has the same effect on all goods and services oil is used to produce or deliver. As a result, if the unit price of oil increases (and there are no changes in money supply, pre-purchase distribution, consumer preferences, etc.) so does the unit price of most consumer goods and services, with the quantity of goods traded likewise dropping.

What does not happen is the same basket of goods being traded, with those impacted by oil getting more expensive but that being offset by other goods getting cheaper, leaving overall nominal price levels the same as before despite shifts in per-sector prices. That would be expected if all non-oil goods had perfectly inelastic supply, but not otherwise.

> Nowhere did I say that supply was inelastic

No, you just described an effect that is total nonsense unless you assume perfectly inelastic supply.

> now you only have $3 to spend on other things, which reduces the price of those other things.

That's only true if anyone is willing to sell those other things at lower prices. With changes in price in something as base-level as oil, the production costs of virtually all goods and most services increase. For low-margin goods, price can't decrease even if demand decreases. As people see their income dwindle, most only buy strict necessities and stop buying other things entirely - leading to increases in the price of luxuries to make up for reduced demand (targeting only those who can afford things at higher prices).

Let's say John has $50 and Sally has $10. The price of X which they both need goes from $5 to $7. The price of Y that they both want but don't need used to be $5. After buying X, John has $43 and Sally has $3. Sally can't afford to buy a Y, but since she doesn't need it, that's ok. The makers of Y realize that they will only get $5 if they only sell to John, where in the past they were making $10. So, they increase the price of Y, knowing that John will still afford to buy it at $7, so they now make $7 in total. If instead they had reduced it to $3, both John and Sally would have bought a Y, but the company would only make $6 in total.

In the 70's, the government tried hard to blame inflation on the oil embargo, greedy capitalists, and the avarice of unions. Gerald Ford's idiotic W.I.N. buttons (Whip Inflation Now) were a particularly pathetic attempt at deflection of the real cause - government running the printing press.

Other notable examples of inflation in America:

1. The "not worth a Continental" inflation in the Civil War.

2. The inflation of the Confederate dollar as the Rebel treasury was a printing press

3. Inflation due to the flood of gold coming from the California gold rush

There is also the inflation in Spain as they looted the New World of gold and silver. It all didn't help Spain that much, as it just produced inflation. Spain never realized what the real wealth in the New World was.

> In order for prices to rise generally, there has to be an increase in the money supply relative to the goods and services it represents.

That's not true, even with simplistic models like supply and demand curves. Prices can rise across the board with a fixed money supply, and you'll have a corresponding reduction in amount of goods traded.

Yesterday 1000 people each bought 2kg of potatoes at $1/kg, and 1 kg of meat at $10/kg. Today, potatoes cost $2/kg, and meat costs $15/kg. So, 800 people buy 3kg of potatoes and 0kg of meat, and only 200 people keep buying both potatoes and meat. Meat producers adjust their production down to account for the new conditions, firing extra staff and sending the spiral ever downward.

It would be nice if the era of economic theory were to end.

Herbert Weisberg argues in Willful Ignorance (paraphrasing for brevity; read and come to your own conclusions) we’re using statistical and probabilistic tools in economics that were purposely left crude by the mathematicians that defined them, as they were defined to handle abstract quantities.

Applying them to human economics means we’re mathematically accepting rounding errors that leave people behind.

That is the pattern we have to break politically. We spend a lot of time emotionally promoting certain stats and hand waving away at a number like “2% reduction in poverty” still means we’re leaving millions behind.

Not having services like universal healthcare is clearly politically gamed, but we refuse to see it as anything like a government bombing civilians.

Reply on news.ycombinator.com