3 ms·
>The economic growth since 1978 was, for the most part, produced by the efforts of 10% of the population I don't see how that could be true. productivity gain
by lukesandberg 15y ago
>The economic growth since 1978 was, for the most part, produced by the efforts of 10% of the population
I don't see how that could be true. productivity gains have been experienced across all levels of the income scale so surely that growth is attributable to more than just the top 10%. Just because that's where the most income growth has been doesn't mean that that is where the growth has been produced, its just where it has been captured.
I think you are right in questioning what drives the inequality. I do believe that some people are more productive and thus deserve additional income and wealth. But i am not so sure that the amount of additional income that the very wealthy earn is actually proportional to their intrinsic productivity. The issue is that capital tends to create more capital because you can 'put it to work' so the wealthy tend to become wealthier. So the rich will naturally become richer because they have access to greater growth opportunities (though investing for instance they can capture growth produced by other companies).
Capitalism fundamentally favors people with excess capital. This is what makes it very effective and efficient because it encourages risk and investment. however it doesn't necessarily produce a fair system because the owners of capital have many means to capture and retain the returns of investment.
For example if in the proverbial pin factory innovations were made that allowed workers to be more efficient (say by specializing different aspects of the production of the pins) then all of a sudden the factory is more productive relative to the labor invested. Now the question is who should capture the gains of that productivity. Should it be the factory owner because he/she put up the initial investment, the factory workers because they are now more productive, the person who came up with the innovation? It's not really clear but i think most people would agree that they all deserve some share of the additional profits. The shift over the last 40 years has been that mostly its going to the owner (the capital investor) and the innovator and not the worker. It's tough to determine exactly how the profits should be divided fairly, because fairness is so complicated in this instance. This is one of the major benefits of labor unions because they provide a way for the workers to share in the additional wealth that they are helping to create. Now obviously not all occupations should have unions and not all unions are effective, but this is the problem to which they are the solution, but you have to admit that its a problem first.
- babblefrog 15y agoI suspect that supply and demand have a lot to do with it. Capital and innovation is scarcer than labor, especially unskilled or lower skilled labor.
- Hyena 15y agoDiffusion, not innovation, drives economic growth. This is simple to demonstrate by looking at the rate of catch-up growth. If there is a problem with economic growth, it is more likely that there are too few diffusion pathways than too little innovation. Our patent system would tend to point in the direction of diffusion problems, no?
- lukesandberg 15y agoyeah that is definitely true, but the underlying issue is fairness (which is trickier).
- yummyfajitas 15y agoI don't see how that could be true. productivity gains have been experienced across all levels of the income scale so surely that growth is attributable to more than just the top 10%. This statement betrays a misunderstanding of how productivity is defined. There is no macro-level measure of individual productivity, so it's simply not possible to determine which levels of the income scale experienced productivity growth. Productivity is only measured, in the most granular cases, at the organizational level. The data available simply cannot determine which workers are more productive. I'll give you an example. My startup involves a lot of categorization - we get a data feed full of poorly labelled fashion items, and we have a team of 6 girls categorizing them (item 1 -> "Handbags", item 2 -> "Dresses", etc). At the macro level, my organization recently increased categorizing productivity by 1000% (i.e., we categorize 10x more items than before). This didn't happen because my girls suddenly started clicking 10x faster. This happened because I wrote a program which can parse item names ("Marc by Marc Jacobs sleeveless shift dress" -> "Dresses") with about 90% accuracy [1]. Macroeconomic statistics have no way of knowing which worker to attribute this productivity gain to. [1] Categorizing natural language is surprisingly hard - stuff like "dress shoes" and "swimwear top" confuses the matter tremendously. Some items also get names like "Cherry Pie"...