--- schema_version: "library-reader-v1" source_type: "original_book" title: "AI经济：机器人时代的工作、财富和社会福利（被科技洪流裹挟的每个人，都需要在AI时代做出自己的选择）" author: null edition: null publisher: null year: null isbn: "9787213100970" library_document_id: "047e618c-de46-43fe-9dc3-84a4dab9fd04" source_sha256: "8d22343cb9934604eebf143b3063c9568282fe03a946d04e2da3dd693ba3f20c" pdf_physical_page_count: 514 pdf_physical_page_index_base: 0 range_start: 480 range_count: 10 html_url: "https://workshop.closeclaw.top/api/ai/read/047e618c-de46-43fe-9dc3-84a4dab9fd04?start=480&limit=10" markdown_url: "https://workshop.closeclaw.top/api/ai/read/047e618c-de46-43fe-9dc3-84a4dab9fd04.md?start=480&limit=10" previous_url: "https://workshop.closeclaw.top/api/ai/read/047e618c-de46-43fe-9dc3-84a4dab9fd04.md?start=470&limit=10" next_url: "https://workshop.closeclaw.top/api/ai/read/047e618c-de46-43fe-9dc3-84a4dab9fd04.md?start=490&limit=10" --- # AI经济：机器人时代的工作、财富和社会福利（被科技洪流裹挟的每个人，都需要在AI时代做出自己的选择） > 页码规则：PDF_PHYSICAL_PAGE_INDEX 是从 0 开始的权威文件物理页索引；PRINTED_PAGE 是页面中识别出的印刷页码，可能为空，二者不得混用。 ## PDF_PHYSICAL_PAGE_INDEX=480 | PRINTED_PAGE=NOT_DETECTED 第1章 [2] Gordon, R. (2012) Is US Economic Growth Over? Faltering Innovation Confronts Six Headwinds, Working Paper: August, Massachusetts: NBER. [3] Actually, in his book published in 1817, David Ricardo warned that, for a time at least, the new technologies of the industrial Revolution could make workers worse off. According to Paul Krugman, modern scholarship suggests that this may indeed have happened for several decades. [4] Morris, I. (2010) Why the West Rules–For Now: The Patterns of History, and What They Reveal About the Future, New York: Farrar, Straus and Giroux, p. 492. [5] Critics will, I am sure, chastise me for showing the simple numbers in this chart, rather than using a log scale. In truth, I agonized over this choice, and also considered showing rolling 100- year increases. But the picture would not be that different, and in keeping with the book's objective of being readily accessible to the general reader, I was loath to introduce such things as a log scale, which could easily put off many readers. So simplicity won out. [6] As you might expect, these figures are the subject of some controversy among economists. They come from the famous work of the economist Brad De Long, "Estimates of World GDP, One Million B.C.– ## PDF_PHYSICAL_PAGE_INDEX=481 | PRINTED_PAGE=NOT_DETECTED Present", 1998, http://econ161.berkeley.edu/. They include the estimated benefits of new goods. (This is known in the literature as the "Nordhaus effect", after the economist William Nordhaus.) De Long also shows figures that exclude these benefits. The figure for the year 2000 is then only about 8½ times the figure for 1800. [7] There is a useful summary and discussion of the long historical trends in GDP and productivity in a Capital Economics study Vicky Redwood and Nikita Shah (2017) History Does Not Suggest Pessimism about Productivity Potential, Capital Economics, November, https://research.cdn-1.capitaleconomics.com/f993f5/history-does-not-support-pessimismabout-productivity-potential.pdf. [8] See Mokyr, J. (1990) The Lever of Riches, New York: Oxford University Press. [9] See Williams, T. (2003) A History of Invention from Stone Axes to Silicon Chips, London: Time Warner. [10] Scott, J. (2017) Against the Grain: A Deep History of the Earliest States, New Haven: Yale University Press. [11] This global picture looks slightly different for some key countries–but only slightly. The USA experienced some reasonable growth in per capita GDP in both the seventeenth and eighteenth centuries. The UK even managed growth in per capita GDP of as much as 0.3 percent per annum from the sixteenth century onward. But even in these two cases the growth rate paled into insignificance beside what was to come later. [12] Malthus, T. (1798) An Essay on the Principle of Population, London: J. Johnson. [13] Ibid. [14] Darwin, C. (1868) The Variations of Animals and Plants under Domestication, United Kingdom: John Murray. ## PDF_PHYSICAL_PAGE_INDEX=482 | PRINTED_PAGE=NOT_DETECTED ## PDF_PHYSICAL_PAGE_INDEX=483 | PRINTED_PAGE=NOT_DETECTED [25] Admittedly, if you look at the recent growth figures for the world as a whole, things do not look too bad. In the years from 2008 to 2016, the growth of global GDP per capita averaged 2 percent per annum. This is slower than in the Golden Age from 1950 to 1973 and the period in the early 2000s when the emerging markets were roaring ahead, led by China. But it is faster than all other periods since 1500. Yet this gives a totally misleading impression that all is well. The emerging markets have continued to grow pretty decently, albeit at much slower rates than before. When you look at the developed countries on their own, though, you see a very different picture. Indeed, in much of the developed world, productivity growth has fallen to almost zero. Since 2008 the growth of GDP per capita in the USA has been 0.6 percent, the lowest since the 1600s. In the UK it has been 0.4 percent, the lowest since the eighteenth century, and in Sweden it has been 0.7 percent, the lowest since the early nineteenth century. The British economists Nicholas Crafts and Terence Mills have estimated that in the early 1970s US Total Factor and Productivity (TFP), that is to say output per unit of capital and quality-adjusted labor (which is a measure of innovation or increased productivity that is not simply due to more factors (e.g., capital) being employed), was growing at just above 1.5 percent. It is now growing at about 0.9 percent. See G. Grossman (2018) Growth, Trade and Inequality, Econometrica, 86(1): pp. 37–8. [26] As we saw in our brief review of ancient history above, it is not sufficient to generate rapid overall productivity growth for there to be rapid productivity growth in one sector of the economy. What also matters is the rate of productivity growth in the other sectors of the economy into which labor is released from the sector experiencing rapid productivity growth. So, it is possible that one sector, say IT, experiences rapid productivity growth and so do the sectors employing a good deal of IT, but if the labor released by the employment of IT systems finds employment in restaurants and care homes, where both the level and the growth rate of productivity is much lower, it is quite possible for the overall growth of productivity to fall back. This was the insight of the American economist William Baumol, presented in an article published in the 1960s. W. Baumol (1967) The Macroeconomics of Unbalanced Growth, American Economic Review, 57(3) (June): pp. 415 –26. ## PDF_PHYSICAL_PAGE_INDEX=484 | PRINTED_PAGE=NOT_DETECTED [27] But, of course, such a result is not bound to occur. It is more likely to occur the lower the sensitivity of demand for the output of the sector experiencing rapid productivity growth to lower prices (which will increase the amount of labor released into the rest of the economy), and the lower the rate of productivity growth in the rest of the economy. If the technological advance is big enough and its effects spread generally throughout the economy (as with electricity, for instance), then this negative result is unlikely to happen. Accordingly, believing that it is happening now when it didn’t in the first several decades of the twentieth century effectively amounts to the proposition that current and recent technological changes are simply not as significant as earlier ones. So the "Baumol" interpretation of the current technological slowdown really amounts to a subset of the school of economists who espouse "technological pessimism," that is to say, those who believe that, in the wider sweep of history, recent and current technological developments don’t amount to very much. [28] Gordon, R. J. (2016) The Rise and Fall of American Economic Growth, USA: Princeton University Press. [29] Solow, R. (1987) We'd Better Watch Out New York Times Book Review, July 12, 1987. [31] Feldstein, M. (2015) The US Underestimates Growth, USA: Wall Street Journal, May 18, 2015. Mind you, not all economists agree. A study by Byrne, D., Oliner, S., and Sichel, D. concludes the exact opposite. They reckon that the effect of correcting mismeasurement was to raise TFP growth in the tech sector and to reduce it everywhere else, with next to no net effect on the economy overall. See Bryne, D., Oliner, S., and Sichel, D., Prices of High-Tech Products, Mismeasurements, and Pace of Innovation, Cambridge, MA, National Bureau of Economic Research, 2017. ## PDF_PHYSICAL_PAGE_INDEX=485 | PRINTED_PAGE=NOT_DETECTED ## PDF_PHYSICAL_PAGE_INDEX=486 | PRINTED_PAGE=NOT_DETECTED 第2章 ## PDF_PHYSICAL_PAGE_INDEX=487 | PRINTED_PAGE=NOT_DETECTED [10] For a discussion of machine learning, see Craig, C. (2017)"Machine Learning: The Power and Promise of Computers that Learn by Example," London: The Royal Society, https://royalsociety.org/\~/media/policy/ projects/machine-learningreport.pdf. ## PDF_PHYSICAL_PAGE_INDEX=488 | PRINTED_PAGE=NOT_DETECTED [21] Max Tegmark has laid down three criteria for judging whether a job is more or less likely to be challenged, or replaced, by robots any time soon. They amount to essentially the same as McKinsey's two criteria, with my suggested addition of "common sense." They are: Does it require interacting with people and using social intelligence? Does it involve creativity and coming up with clever solutions? Does it require working in an unpredictable environment? Tegmark (2017), p. 121. ## PDF_PHYSICAL_PAGE_INDEX=489 | PRINTED_PAGE=NOT_DETECTED 下一段：[继续读取](https://workshop.closeclaw.top/api/ai/read/047e618c-de46-43fe-9dc3-84a4dab9fd04.md?start=490&limit=10)