I do. Pull away. I got my twelve gauge, bandoleer and tinned dog food…and I got your back!
Yuriy, I think this is a good point and it’s something I’ve thought myself.
However, I think it’s important to make some clarifications. What is the maximum level of accuracy in consumption planning that could be achieved by any system? The fundamental limit here is ‘primary uncertainty’, i.e. the limits of human knowledge itself. On the face of it, consumption planning for a single person can be no more accurate than that person’s ability to project their preferences into the future (e.g. 1 year). If that person is not able to predict that they will turn vegetarian in 6 months, then that consumption cannot be planned.
My point is that we cannot exceed the limits of human knowledge itself. The only way to entirely overcome this uncertainty is for a central agency to tell consumers what they’re going to have - but that’s a hollow victory. Unless there is a way to perfectly predict what each person will consume for the next year then there will be some error and some forecasting will be required. Also some adapting through the year.
The next questions are: (1) Realistically what can be achieved? (2) Exactly what are we trying to achieve by accurate consumption planning?
Addressing (1), of course there are things the individual might not be able to predict which the system itself can. An individual might not know they will become vegetarian in 6 months, but statistically there will be X number more vegetarians in region Y based on current trends. I think forecasting would be much better in a parecon because planning support units would have free access to aggregate consumption data and could much more efficiently process it. In a capitalist society different parts of the system are segregated and don’t freely share information.
I think the planning categories will be constantly refined in response to experience. Planning specialists will compare consumption proposals made at the beginning of the year with what was actually consumed at year’s end. They will interpret how the initial proposal connected to reality. Then the planning procedure, including the categories and the consumption proposal interface, will be slightly altered so that proposals better match real consumption. This would basically be a process of more effectively mapping a user-friendly consumer interface onto a detailed list of real use values made by production units.
I’ve said before that I think consumer federations should be able to alter consumption proposals at aggregate levels if the individual consumer proposals fail in some regard (e.g. undersupply some good in a way that is an error rather than reflecting real consumer preferences). That is because the consumer federations have access to aggregate data, e.g. toaster sales for the past 10 years. Again, this will vary over time according to experience. The aim would be for the planning process to be refined so that this was less and less necessary.
I have a lot more to say on this but this comment should be no longer.
As a thought experiment, I’m trying to imagine what would happen if I was presented a complete catalogue of 100 million distinct products on Parezon.com and you gave me 2 weeks 24/7 to fill out my yearly consumption proposal. Could I provide a list of what I would consume, in concrete terms, for the next year? No.
However, with assistance I could at least do a lot better. Firstly, accessing last year’s consumption list would help a lot. Then I could use that as a template.
I think it’s helpful to consider what we actually consume. Here is a screenshot from this article about the average expenditures of a US household [US Bureau of Labour Statistics, 2020 data]. It doesn’t matter if it is exact, it gives a rough idea of the pattern.
Firstly, I can see there are some categories there which are more suitable than others for specifying in advance. For example, ‘apparel and services’. I would not like to pick out all my clothes through a screen. I would want to try them all on first, or at least eyeball them, with rare exceptions. (Now maybe this could be partly overcome with some kind of Lidar scan, but I digress!).
Similarly, ‘restaurants and other meals away from home’. I’m not going to plan out every restaurant I will eat at through the year. That’s just not how restaurants work. ‘So what do you feel like eating?’ ‘Well, I planned for 8 meals at Siam Thai, so let’s go there!’.
So in these type of cases, it doesn’t make sense to even try to plan for actual use values. They should be no more granular than subcategories.
Secondly, from that list, housing, transportation, personal insurance / pensions, healthcare - the top 4 expenses - and education, are all either matters primarily for collective consumption or are more relevantly addressed through long term planning.
Where do more exact consumption proposals make sense? Firstly, where people have very exact tastes. For example ‘tobacco/smoking products’. I could see myself ordering a specific packet of pipe tobacco for the following year, e.g. ‘50g Peterson Irish Flake X 3’, not just the generic subcategory ‘pipe tobacco’. Secondly, it depends what you mean by ‘exact consumption proposal’.
I think it’s worth considering the detriments of overly coarse categories versus the detriments of inputting incorrect but exact information. I may know I will buy a computer with roughly XYZ specifications in a budget of 1500-2000 this year. If I force myself to pick an exact make and model from the product catalogue, yes the information is more exact. If this turns out to be correct, then helps the producer. If this turns out to be false, this is sending unhelpful noise to the producer. A coarser choice must be interpreted / extrapolated by producers but might end up sending less noise.
Again, I have a lot more to say but I will leave it here. I will finish by saying that like everything it is best to make a hierarchy of priorities. It is most important to first cover the biggest and most essential goods/services. As you work down that hierarchy, the stakes are lower and lower.
That’s a very useful table on average expenditure for a US household - thanks for sharing. I’d be curious to know the method they used to get the data. If we are able to start work on building the par-planning interactive prototype this would be great to use as a starting point for a consumer in the annual plan.
What would be really interesting is to have this kind of data over longer periods of time and be able to measure changes in household expenditure for different categories of goods and services. I don’t know if this data exists, or how it could be gathered, but if not someone should do a Phd research project on it.
The data comes from the US Bureau of Labour Statistics:
They track changes over time and how different categories fluctuate.
These are great discussions…for me ar least. Must admit my favourite comment was…
“(Now maybe this could be partly overcome with some kind of Lidar scan, but I digress!).”
Indeed.
I’ve always thought, ok, participatory planning may be partly flawed, imperfect, but what’s the alternative? Part planning, part markets? All markets? I started to consider what would happen if consumer advocacy…you know, those movements, groups, government departments or projects that help consumers with info about products, what goes in them and stuff, how they’re produced etc., those who work to expose bad workplace practice in all aspects, and on….was taken to its extreme. In other words, toward a society that ‘produced’ an actual knowledgeable and rational consumer. To me, this gets lost in all this. Market capitalist economies by nature require irrational consumers. That’s why they only research x amount and the advertise later to sell and increase market share. It’s about keeping the consumer dumb. But if everything about what is being produced and how, was or COULD BE KNOWN, and there were NO workplaces peddling bullshit re products and services (advertising agencies and corresponding market research workplaces designed merely to enhance advertising nonsense), then the ‘market’ would be undermined, as would various institution imperatives that drive capitalist laws of motion.
In a certain sense, consumers with power to access info regarding all products if necessary, whether some workers are being exploited and alienated to the nth degree, if they wish to, would be in a position to dictate production directly, even specifically but also INDIRECTLY. But in some sense people don’t need to be specific about their consumption for the advancement and fostering of shared values, which is the real foundation of Parecon, because it’s not about whether it’s allocation system is a perfect planning system. What’s more important is that they are now effectively effective and affective consumers AND workers/producers themselves, in companies, workplaces whose products and practices have been exposed and stored somewhere…known or can be known. This is anathema to market imperatives. Markets by nature keep hidden as much as can be…they lie on purpose…through whatever means capitalists can find…to ‘produce’ IRRATIONAL consumers. Parecon works through integrated institutional structures operating together to foster certain values, producers, and rational consumers who have all been schooled on such matters.
The basic thing to realise is there will be trade offs across the economy in lots of areas, not just planning, I guess. It’s whether those trade offs are worth it. Planners in a Parecon would be rational workers and consumers, all workers would be consumers and consumers workers, who are not just planning for consumption but planning for leisure and time worked. In fact, their wage is really merely their individual consumption plan, what they take from the social pie. That’s how much they will earn, on average, for that year, along with everyone else, excepting those who are saving or putting themselves in debt, which dictates how much they will have to work…a quid pro quo.
Even if their individual plan is merely “what I consumed last year” and nothing more, and everyone else did the same, there will be a process (hypothetically leaving out all specifics, nuance and ecological stuff), that eventually arrives at some acceptable work/leisure balance. Consumers as workers don’t want to be run off their feet making shit products for a bulk of people that they know are crap for everyone and the ecology. The idea that ecological stuff and other info re specific products and services would not be included in that process, through information coming from those more conscientiously informed and concerned consumers and producers, would be insane. That alone would be infinitely better than what we have now, regardless of the courseness of the planning.
So it won’t be perfect, that’s a given, and it may even appear to some that participatory planning is still kind of like markets in ways…forecasting, but it would now be rationally done. Now economic courses could be done that can talk about rational consumers for real and their predictions would/could be more correct. But being kind of like something isn’t being it. Some say music is a language, but it isn’t. It may have qualities like a language (doubtful) but if it was a language it wouldn’t be like one. But it’s not even whether x is something or isn’t or what it’s called. Is it really planning or not really? It’s about whether it produces, fosters, the desired values and ecological sanity.
So while these discussions are of value, and they are, at least for me, it must be not lost what Parecon is trying to achieve. A participatory planning system doesn’t have to be a perfect planning system or even a planned system by definition, necessarily. It’s merely (well, not really merely merely), has to achieve what markets cannot and what consumer advocacy type projects exist to correct.
Dear @AfterTheOligarchy ,
I asked Paul Cockshott this very question (how many different goods/services) in an interview I did with him recently
thank you for this interesting link! Paul Cockshotts estimate was the only quantitative estimate in the number of products in a society i have seen so far. I really appreciated that and would like to return the favor by pointing out a very new recent estimate. Its different but related, as it doesnt estimate products but measures (!) firms in hungary.
Hence, it gives us an idea of how many councils might be necessary to reproduce a economy of say hungary.
In a recen Nature publication(https://www.nature.com/articles/s41598-022-11522-z#Ack1) the CSH Vienna used VAT data of the Hungarian Nationalbank to reproduce the entire Hungarian supply-chain. It involves 91.595 Companies.
Now, im not sure how many products an average company produce, but once we have such an estimate we might also be able to better estimate the number of products to organize in total.
1 Like
Hi @ccc,
Glad you found it interesting. More where that came from!
Thanks for the link to the paper. Looks interesting. Skimming through and looking at the authors seems to be a complex systems approach which piques my interest. But that figure you quoted is interesting. With a population of ~10 million people in Hungary, that means approximately 110 people per firm.
I’d be very interested in the distribution of firms. How many are big, medium, and small. That would be important info for the participatory planning procedure. My reasoning is that it’s most important to get the proposals of large firms to converge because they account for greater supply. That way you can reduce the number of iterations in annual planning.
(Of course, the distribution of sizes in a parecon would not necessarily be the same as a capitalist economy, but it’s a starting point).
Actually, if we allow ourselves to speculate a bit here (back of the napkin stuff), if there were let’s say (guessing) 100 million distinct products being sold by 90,000 firms in Hungary, then that would be roughly 1000 distinct products sold per firm on average. If there were 1 billion products, it would be 10,000 products per firm on average; if 10 million products, then 100 products per firm on average.
I’ll give the paper a proper read when I get the chance.
Here is a link (not an academic paper) to a webpage of statistics about Amazon.com: How Many Products Does Amazon Carry? - Retail TouchPoints. They assert that Amazon sells 350 million distinct products (“not including books, media, wine, and services”). I’d like to see a proper citation but it’s not a surprising figure either. It’s in the 0.1 - 1 billion range which I’d expect.
This is a treasure trove:
The North American Industry Classification System (NAICS). Classification of the economy into sectors, subsectors, etc.
(It’s really astonishing to think how much work over the decades it must have taken to develop this. Imagine handing that to someone in the 19th century.)
This is a quick overview webpage: NAICS Codes & Understanding Industry Classification Systems
P.S. I stumbled across something called IMPLAN which seems to be some kind of commercial planning software which uses input-output tables … e.g. https://support.implan.com/hc/en-us/articles/360043232993-Industry-Aggregation, also https://support.implan.com/hc/en-us/articles/115009505587-Key-Assumptions-of-IMPLAN-Input-Output-Analysis. Could be interesting, or no use. Linking here just in case.
In USA, Canada, Mexico they used NAICS. In Europe, it’s NACE:
And the Combined Nomenclature (CN):
https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L:2021:414:FULL&from=EN
Jason,
What would be really interesting is to have this kind of data over longer periods of time and be able to measure changes in household expenditure for different categories of goods and services.
Go to this link, download the Excel spreadsheet and look at tab 4.2. They show the % of household income spent on 12 different categories of goods/services from 2001-2020.
1 Like
This is really useful data, thanks. Well done for finding it!
Interesting to see it confirmed in the data that in that twenty year period household expenditure changes very little - the 2021 pandemic year being the only outlier.
This would be a great starting point for us to use in building a participatory planning software prototype.
1 Like