I do think there are many things that USSR could have done differently. It was a huge country and could have been largely self sufficient given that DPRK managed to survive being much smaller and having far less advantages.
In my view, the problems come down to the way selection process in the party worked, which allowed people like Khrushchev and Gorbachev get to the top. Khrushchev destroyed relations with China which facilitated American detente in the 70s and further isolated USSR. If The Soviet Union and China combined powers instead, we’d be living in a completely different world today.
Another example would be shutting down of Artels (Артель) which provided a structure of many small-scale producers in USSR that would fill the niches like belt buckles, specific forms of toys and other minute necessities that can be fulfilled more efficiently without central planning. Such a worker formations worked for themselves, were prohibited from growing beyond a certain size and had to prove they had no exploited workers, but they allowed to have more flexibility in the market.
Then there was OGAS which could have been implemented in the 70s, and would’ve been completely transformative. Incidentally, it does look like China is moving in that direction. Their use of digital currency is a step towards facilitating that. This was a surprisingly good read on the subject. Once you have digital tracking integrated into the supply chain, you have real-time information about the state of all the pieces in the supply chain. And nowadays, we have stuff like AI systems that can do dynamic allocation better than markets can. While in the west, most of the focus with this tech has been on making chat bots, in China it’s being actively integrated into industry and supply chains where it will provide immense value.
Overall, the big insight Stalin had with Artels and CPC has embraced was that markets can be used as allocators within central planning. One of the core weakness of the Soviet model is that the planners need to have accurate and up to date information to make sound decisions. But the problem is that you inevitably have a lot of noise in the system. There are honest mistakes, corruption, lack of transparency or inadequate data collections, and so on. The data also takes time to be collected, aggregated, and analyzed. So, by the time the plans are made, they are based on incomplete data that’s also out of date. Then it takes more time for the plan to be implemented, and thing get more and more out of sync as the process goes on. The tendency ends up being to overcorrect, which further compounds the problems.
Chinese approach is to avoid doing micromanagement. The central planners set general direction, and provide incentives, then let the markets do dynamic allocation at each level. In this scenario, planners simply have to figure out how to create the right selection pressures to direct labor and resources in the desired direction. Support companies that do things that fit with the plan, pass legislation to suppress enterprises that go counter to it, and let things work themselves out.
DeepSeek is a really good example of this approach in action. The party decided that quant trading was harmful and banned it. High-Flyer was one of the hedge funds involved in it, and they saw the writing on the wall. They realized they had a bunch of GPUs and math talent on hand, so they pivoted to AI, and ended up producing one of the best models around. The government never had to tell them to do any of this explicitly, they simply created the conditions for it.
I do think there are many things that USSR could have done differently. It was a huge country and could have been largely self sufficient given that DPRK managed to survive being much smaller and having far less advantages.
In my view, the problems come down to the way selection process in the party worked, which allowed people like Khrushchev and Gorbachev get to the top. Khrushchev destroyed relations with China which facilitated American detente in the 70s and further isolated USSR. If The Soviet Union and China combined powers instead, we’d be living in a completely different world today.
Another example would be shutting down of Artels (Артель) which provided a structure of many small-scale producers in USSR that would fill the niches like belt buckles, specific forms of toys and other minute necessities that can be fulfilled more efficiently without central planning. Such a worker formations worked for themselves, were prohibited from growing beyond a certain size and had to prove they had no exploited workers, but they allowed to have more flexibility in the market.
Then there was OGAS which could have been implemented in the 70s, and would’ve been completely transformative. Incidentally, it does look like China is moving in that direction. Their use of digital currency is a step towards facilitating that. This was a surprisingly good read on the subject. Once you have digital tracking integrated into the supply chain, you have real-time information about the state of all the pieces in the supply chain. And nowadays, we have stuff like AI systems that can do dynamic allocation better than markets can. While in the west, most of the focus with this tech has been on making chat bots, in China it’s being actively integrated into industry and supply chains where it will provide immense value.
Overall, the big insight Stalin had with Artels and CPC has embraced was that markets can be used as allocators within central planning. One of the core weakness of the Soviet model is that the planners need to have accurate and up to date information to make sound decisions. But the problem is that you inevitably have a lot of noise in the system. There are honest mistakes, corruption, lack of transparency or inadequate data collections, and so on. The data also takes time to be collected, aggregated, and analyzed. So, by the time the plans are made, they are based on incomplete data that’s also out of date. Then it takes more time for the plan to be implemented, and thing get more and more out of sync as the process goes on. The tendency ends up being to overcorrect, which further compounds the problems.
Chinese approach is to avoid doing micromanagement. The central planners set general direction, and provide incentives, then let the markets do dynamic allocation at each level. In this scenario, planners simply have to figure out how to create the right selection pressures to direct labor and resources in the desired direction. Support companies that do things that fit with the plan, pass legislation to suppress enterprises that go counter to it, and let things work themselves out.
DeepSeek is a really good example of this approach in action. The party decided that quant trading was harmful and banned it. High-Flyer was one of the hedge funds involved in it, and they saw the writing on the wall. They realized they had a bunch of GPUs and math talent on hand, so they pivoted to AI, and ended up producing one of the best models around. The government never had to tell them to do any of this explicitly, they simply created the conditions for it.