The macro frontier (not your grandpa’s Keynesian economics, or is it?)

We argue that the input-output network of investment goods across sectors is an important propagation mechanism for understanding business cycles. First, we show that the empirical network is dominated by a few “investment hubs” that produce the majority of investment goods, are highly volatile, and are strongly correlated with the cycle. Second, we embed this network into a multisector model and show that shocks to investment hubs have large aggregate effects while shocks to non-hubs do not. Finally, we measure realized sector-level productivity shocks in the data, feed them into our model, and find that hub shocks account for a large and increasing share of aggregate fluctuations. This fact allows the model to match the decline in the cyclicality of labor productivity and other business cycle changes since the 1980s. Our model also implies that investment stimulus policies increase employment throughout the economy but have unequal effects across sectors.

That is from a new NBER working paper by Christian vom Lehn and Thomas Winberry.


"vom Lehn" but not "von Lehn"? I never saw this "vom" before though my knowledge of German is a bit limited.

Origin Danish? Frisian? Dutch?

Delete "Dutch": that would be "van" presumably.

It means Orange Man Bad in Pelosi-speak.

Ok after googling this it seems like "vom" is short for "von dem". Still quite unusual to see this in a surname i would say.

Do business cycles even matter anymore? In the last few decades, the asset market bubble blowing and the closely related central bank shenanigans matter more than anything else in the economy.

what matters is how woke your physics is

The Trump tax cut was sold on the promise that it would stimulate an enormous investment in productive capital, thus triggering an upward phase of the business cycle. Not. It's been two years since Cowen and Tabarrok linked their videos on the various theories of the business cycle. Let's do it again.

Tax cuts only resulted in more stock buybacks. Wages haven't budged despite unemployment numbers. Trump's own removal of SALT deductions made him now a Florida resident. Easy for the rich to dodge but not so easy for the middle class who once again is stuck with the bill.

"Wages haven't budged despite unemployment numbers. "

That's wrong. Obviously wrong.

"Wages for U.S. workers grew 4.0 percent over the last year, increasing the average wage level by $1.09 to $28.54 an hour, according to the ADP Research Institute® Workforce Vitality Report (WVR) released today.
Since January 2019, female job holders received average wage gains of 5 percent, while men averaged wage gains of 4.6 percent.""

Wage growth has gone from 2.5% in the 2016-2017 period to over 4% as of 2019. That's a 80-100% increase in wage growth in the last 3 years.

ADP Research Institute's data is inconsistent with BLS data.

The BLS data still proves wages have gone up. The ADP data is in nominal rates, whereas the initial BLS data is in real rates.

From the BLS data you linked: "Real average hourly earnings increased 1.1 percent, seasonally adjusted, from November 2018 to November 2019. "

Also from the BLS data: Nominal over the year percent changes Oct 2019 = 3.5%

True, wages have gone up but 3.5% < 4% and real wages matter (3.5%-2.1% = 1.4%). There was growth, but not robust as your post suggested.

I was responding to the claim: "Wages haven't budged despite unemployment numbers. "

That claim was wrong. That was the point of my post.

Abstract tree theory, folks.

The constant rescaling our economy does because we deal with finite, Newton's grammar is only so good then we have to solve the quantum system.

I suspect these folks referenced Krugman and the agglomeration thesis. Abstract tree, flow is forced through constricting channels (Tree trunks) and is thus must keep the 'basket' size optimal such that the tree trunk has enough congestion to execute a commutative property and resolve loops via the distributive property up the agglomeration network, yielding a nearest approximation to Newton's continuous flow. Everything in the universe has to make its grammar works, first, or in Shannon terms, manage the encoding map.

The economists are getting 'woked'.

The mental gymnastics that make it work.

The deja vu effect is a recognition of a semi-repeatable sequence, and we go back through recent history and list the key events between to peaks =, for example, oil price peaks. (This is Hayek) The list is finite, there is an Avogadro's N, and thus an implied uncertainty constant (this is Nash). Everyone has the map (this is Shannon). We can get the map because we can separate transaction costs, move them up our inventory value added chain and aggregate them (This is really Coase). Steve Hawkins was about the need to have enough flow to send the map, in finance this is trigger the shorts. I still suggest we name the process Baumol. We know how to automate this, Google's Go game player uses this model.

OK, what is this stuff about no on likes waiting in line?

That is when we have deja vu, when we get stuck in long lines, like sudden gas shortages. It is the moment when we say,'I could have predicted this", and now you can. We shack our head and when going to bed ponder our discovery getting a does of dopamine, we sleep and the neural nets are weighted via new synapse. Our neutral nets acquire the map, and we plaster he map everywhere, in trade books, professional journals, price lists, etc etc. The breakthrough that makes this work is that information is just one of many, if not all, optimally congested flow problems. The ability of nature to make its own grammar.

High leverage amplifies random noise creating instability in the financial sectors making them more important in driving cycles. The obviousness is apparent to anyone who has looked at system dynamics.

Comments for this post are closed