I was very grateful, and felt very privileged, to appear on the Steve Keen & Friends podcast on Saturday, to present the One Lesson. The production, managed by Ty Keynes, is impressive, and I was made very welcome.
I’d be very interested in any thoughts you have on how it went - please do let me know in the comments here. In the meantime, here’s my own overall impression, followed by some comments on what I could have done better.
It got off to a good start, with the dominoes on a grid puzzle, illustrating how invariants link the micro scale to the macro scale. I thought Dan Sanderson in particular recognised the importance of this.
The middle bit was very unclear, where I tried to respond to Steve’s very specific questions about modelling depreciation, the environment and entropy before I’d properly set out the model for the audience.
The more general chat towards the end was friendly enough, and hopefully gave a bit of context to my thinking.
The main point I would have liked to clarify is how actions preserve raw net worth1. It’s not that the value of something which you buy is equal to the money you spend on it: it’s that, for example, your spending decreases your RNW by exactly as much as the seller’s increases, and similarly that what you receive in return increases your RNW by exactly as much as the seller’s decreases. Each action (other than production and consumption) preserves the combined RNW of the 2 participants.
Detailed commentary
Equity
(37:18)
If I’ve understood right, Steve Keen’s Minsky software uses the term “equity” to mean “net financial assets”, or debt assets minus liabilities. This is why the sum of everyone’s “equity” equals zero: for every debt asset there is an equal and opposite liability.
The One Lesson uses the term “raw net worth” to mean total assets (both debt and tangible) minus liabilities. Instead of adding to zero, the RNW of everyone in the world adds to the total of the world’s tangible assets. Even though RNW isn’t conserved (because production increases one person’s RNW without decreasing anyone else’s, and consumption decreases one person’s RNW without increasing anyone else’s), transferring a tangible asset from one person to another does conserve their combined RNW, as does every action involving debts. (See the post on the 7 economic actions for details).
Depreciation
(43:39)
This is something I wish I’d asked to leave until I’d explained the model more clearly first. It took me years to think about how it could be modelled, and while I think the answer I settled on (treating depreciation as the simultaneous consumption of a less worn-out object and production of a more worn-out object) works, and maintains the linearity, it’s not the most elegant part of the model. I did notice that Dan and some commenters also intuitively saw depreciation as a sort of consumption.
Production functions
(45:33)
Because I hadn’t properly explained the linearity by this point, Mike Radzicki quite reasonably asked how it applied to production. It had been a long time since I’d read about production functions, so I didn’t come up with a good answer, but they relate the amount of output to the factors (labour, land, capital) used in production.
His completely valid concern was that I might be assuming that the amount output in production is just a constant times the amount of input. That would certainly be a bad assumption. Ideally, I’d have explained that the linearity in production is that, for example, if you consume some raw materials in the morning, and some more in the afternoon, the total decrease in your RNW is the sum of the two decreases. And if you produce some output in the morning, and some more in the afternoon, the total increase in your RNW is the sum of the two increases.
More linearity (wage dynamics, etc.)
(47:06)
Steve: “If you have wages rising because of rising employment, and employment rising because of high investment, then you’re going to get a non-linearity”.
This is completely correct. If I’d understood at the time what the issue was, I would have said that the linearity shows up in that, for example, if an employer pays employees a certain amount one month, and pays them some more the next month, the total it’s paid them is the sum of the two amounts.
Environmental issues
(52:55)
Steve was concerned the model didn’t take into account extraction from the earth and waste put back in. These are issues in which he has a strong interest.
I’d argue strongly that it’s possible to separate those concerns from the analysis of how people’s RNW changes, and that the model doesn’t have to include them to be useful. But I have some thoughts about how to represent these ideas, which I’ll address in a future article.
Energy missing from the model?
(1:04:46)
The diagram of Alice producing tables from planks didn’t show energy use. “Everything’s being produced with stuff that already exists inside the economy”.
I’d say that energy is something which does exist inside the economy, and can be represented in the model: a producer can buy fuel2 to be used up in the production process like any raw material, or buy energy as a service.
Stocks and flows
(1:11:55)
Mike talked about the model clearly modelling in terms of stocks and flows.
This is completely correct. Stocks are RNW. Flows are ΔRNW.
Adding feedback loops
(1:14:45)
Mike suggested a missing ingredient is feedback loops, and that I should include them in the model. He recognises that this is where it becomes non-linear.
This is largely my failure to explain the model clearly. The benefit of it comes from being a different sort of model: one which reveals the invariants (like the dominoes always covering as many white as black squares), no matter what people decide to do. As we’ve seen in the articles over the last year, RNW by itself explains so many things, from banking, to Ponzi schemes, to firms, and more.
I’m happy to leave the complex system dynamics models to Mike and Steve. I think the simplicity of RNW, while being completely mathematically rigorous, is its key strength.
Final thoughts
The ability that exists today to set up an alternative media outlet, like Steve Keen and Friends, is amazing. For modest cost in historical terms, almost anyone can broadcast to the world. It’s been remarkable that the thoughts I’ve had can be shared with probably a few thousand interested people from multiple continents, and can be critiqued in real time by well-known and highly-experienced economists.
Thanks again to Steve Keen, Ty Keynes, Mike Radzicki, Daniel Sanderson, and Julia Smith working behind the scenes. I hope it provided some food for thought, and if I come back, maybe I can explain myself a little clearer, as well as showing the essence of superposition in the chain and loop diagrams.
Someone’s raw net worth (RNW) is what they own plus what they’re owed minus what they owe. It is a “heterogeneous” sum/difference, which just means that things of different types are added and subtracted, not monetary “values” which have been assigned to them.
The fuel is produced by the seller by extracting it in some way.