Friday, July 31, 2020

Will computer programs replace newspaper columnists?

See How do you know a human wrote this? by Farhad Manjoo of The New York Times.

In my macroeconomics class, we talk about the types of unemployment. Here is one of them:

Structural-unemployment caused by a mismatch between the skills of job seekers and the requirements of available jobs. One example of this is when you are replaced by a machine.

Excerpts from the article:
"This month, OpenAI, an artificial-intelligence research lab based in San Francisco, began allowing limited access to a piece of software that is at once amazing, spooky, humbling and more than a little terrifying.

OpenAI’s new software, called GPT-3, is by far the most powerful “language model” ever created. A language model is an artificial intelligence system that has been trained on an enormous corpus of text; with enough text and enough processing, the machine begins to learn probabilistic connections between words. More plainly: GPT-3 can read and write. And not badly, either.

Software like GPT-3 could be enormously useful. Machines that can understand and respond to humans in our own language could create more helpful digital assistants, more realistic video game characters or virtual teachers personalized to every student’s learning style.

OpenAI has given just a few hundred software developers access to GPT-3, and many have been filling Twitter the last few weeks with demonstrations of its surprising capabilities"

"Give GPT-3 a natural-language prompt — “I hereby resign from Dunder-Mifflin” or “Dear John, I’m leaving you” — and the software will fill in the rest with text that is eerily close to what a human would produce.

These aren’t canned responses. GPT-3 is capable of generating original, coherent and sometimes even factual prose. And not just prose: It can write poems, dialogue, memes, computer code and who knows what else.

GPT-3’s flexibility is a key advance. Matt Shumer, the chief executive of a company called OthersideAI, is using GPT-3 to build a service that responds to email on your behalf: You write the gist of what you’d like to say, and the computer creates a full, nuanced, polite email out of your bullet points.

Another company, Latitude, is using GPT-3 to build realistic, interactive characters in text-adventure games. It works surprisingly well — the software is not only coherent but also can be quite inventive, absurd and even funny."

Related posts:

McDonald’s Tests Robot Fryers and Voice-Activated Drive-Throughs: Burger giant wants to speed service as competition for fast-food diners mounts

Is Walmart adding robots to replace workers or because it is hard to find workers?

Robot Journalists-A Case Of Structural Unemployment?

Structural Unemployment In The News-Computers Can Now Tell Jokes 

WHAT do you get when you cross a fragrance with an actor?

Answer: a smell Gibson.

Robot jockeys in camel races

Are Computer Programs Replacing Journalists?

Automation Can Actually Create More Jobs 

The Robots Are Coming And It Might Not Be A Case of Structural Unemployment 

Broncos to debut beer-pouring robot at upcoming game

Robots Are Ready to Shake (and Stir) Up Bars  

Thursday, July 30, 2020

In 1834 the British government spent about 40% of the national budget to buy out slaveowners to emancipate their slaves

Three Books on Jamaica: Anatomy of an Uprising: Back-breaking labor, an oven-hot climate, whip-bearing overseers and a planter class eager to exploit slave labor. Jamaica exploded in rebellion—and the empire struck back by Fergus M. Bordewich. He reviews three books in The WSJ. Excerpts:
"Just after Christmas in 1831, the British Empire’s wealthiest island [Jamaica] exploded."

"Hundreds of slaves, having been pushed beyond endurance, attacked hated overseers and their masters’ property. “We have worked enough already, and will work no more,” striking laborers told a pair of plantation owners. “The life we live is too bad; it is the life of a dog.” In all, 145 estate houses were destroyed and many others severely damaged."

"The uprising was soon over, having been weakened by its poor organization and thwarted by the failure of the island’s 300,000 slaves to rise en masse. It was also overwhelmed by the firepower of British troops. Few whites were killed, but the colonial elite’s confidence in its ability to defend itself was deeply shaken. Hundreds of enslaved men and women were killed in battle or summarily executed, some simply because they had attended a Baptist meeting."

"The revolt failed to improve conditions for the enslaved in Jamaica, but it crucially wounded the institution of slavery itself."

"it was only one factor in the ending of slavery, along with surging abolitionism in Britain, an increasingly muscular reform movement in Parliament, and the falling price of sugar, the islands only export crop. But the revolt, he says, “sent an unambiguous message to London that slavery was no longer sustainable—not economically, not militarily, and not morally.”"

"colonial Jamaica was characterized by extreme systemic violence against enslaved people. It was also ruled over by a dissolute planter class obsessed with short-term profits that made it cheaper to work slaves to death and buy new ones than to sustain them into their later, less productive years.

Long before India became the jewel in the empire’s crown, Jamaica was seen to be the “colony that was most indispensable to imperial prosperity,”"

"Jamaica’s rulers, many of them absentee plutocrats living in England, were the richest people in the entire British Empire by the 1770s and “probably had more influence within the British imperial state than any other colonial subjects of George III.” They bought seats in Parliament like baubles."

"In the late 18th century, planters could expect rates of return in excess of 17% annually on their investment. Output continued to grow as slavery became more “efficient”: Between 1750 and 1810, the average productivity of enslaved workers doubled as mistreatment became more calculated and systematic. For countless slaves, life was brutish and short. Cane fields were oven-hot and rat-infested. The labor was back-breaking. When cane was boiled during production, liquefied sugar clung to the skin like searing glue. Workers who arrived late in the fields were commonly subjected to a “slight whipping” of 10 to 20 lashes. Many died of sheer fatigue."

"British statesmen “thought that maximizing labor productivity, as with slavery, and producing goods, such as sugar, at the lowest possible cost for the benefit of a growing consumer class was central to any imperial policy,”"

"Slavery’s defenders, meanwhile, claimed that slaves had in fact been “rescued by British benevolence from the Hobbesian world of African cruelty and superstition.”"

"Revolts would recur every few years through the rest of the 18th century and into the 19th. Yet change was stirring. Less in reaction to the bloodletting than as a result of reformist impulses in England, the foundation was being laid for the political attack on slavery that would come to fruition in 1834. Beginning in the 1760s, reformers began to challenge both the immense power of Jamaican planters and the morality of slavery. Their critique gained traction as they made the case that the planters were determined to impose their tyrannical and “un-British” ways of life on free Englishmen, in part because, it was alleged, they would import Black slave labor into Britain itself.

In 1834, faced with rising public hostility and buffeted by fear of further rebellions, slaveowners and their parliamentary allies persuaded the government to buy them out. Commented one: “If the slavery of our colonies is a sin, it is the sin of the nation, and ought to be redeemed at the nation’s expense.” In all, £20 million was handed out to the empire’s slaveowners, about 40% of the national budget at the time. Full emancipation was completed in 1838. As Mr. Zoellner neatly puts it: “Three centuries of bondage were thus terminated with a dull bargain and a cash payout amid the grind of politics.”"

Wednesday, July 29, 2020

From 1720 to Tesla, FOMO Never Sleeps: The South Sea bubble is the classic story of an investing mania. Are investors today any wiser?

By Jason Zweig. I like articles like this that link economics and storytelling. In fact, I have another blog called Dollars and Dragons: A look at the intersection of economics and mythology. If humans are the storytelling animal (which is the name of a book by Jonathan Gottschall), then it wi worth looking at how stories affect or play out in the economy.

Excerpts from Zweig's article:
"In the summer of 1720, shares in the South Sea Co. and other leading stocks roared to all-time highs as speculators chased instant profits. Ever since, this sudden outbreak of stock trading has been known as “the South Sea bubble.” Even faster than it inflated, it burst—and left us with lessons about human nature that reverberate today.

From July 10 through July 12, 1720, South Sea shares perched at £950, up 650% for the year. Royal Exchange Assurance and London Assurance crested in late August, up an astonishing 1,243% and 4,220% for the year, respectively.

Then, in three catastrophic weeks in September, it all began crashing down. By the end of 1720, these leading stocks had fallen between 81% and 96% from their peak."

"They all were sucked in by a perfect magnetic storm: the rapid advent of newspapers, ready loans at low interest rates, and exciting narratives about technological innovation. Above all was the eternal human desire to be part of the in crowd, or what we today would call FOMO, fear of missing out."

"Bubbles are as old as financial markets. Even in ancient Babylon, commodity prices took sudden leaps and falls that can’t be fully explained by weather or war."

"In 1720 . . That June alone, 88 startups, most of them publicly traded, were launched in London. Many sought to raise £1,000,000 to £5,000,000 apiece (roughly $190 million to $945 million in today’s money)"

"Fistfights broke out over the right to buy stock while it still could be had; speculators thronged London’s financial district to buy shares in any company, desperately pleading, “we don’t care what it is.”"

"New media technologies—newspapers in 1720, radio in the 1920s, the internet in the 1990s, social media and smartphone apps today—are “the cultural substrate in which a mania can grow,” says William Deringer, a financial historian at the Massachusetts Institute of Technology.

As word spreads that “everybody” is doing something, it can be hard for anybody to resist joining."

"Imitation isn’t always irrational, either. It helped our ancestors save mental and physical labor and adapt more quickly to changing environments"

"That can quickly coalesce into a narrative, in which our imaginations rapidly transport us to different places and times. (“I’ll become rich, just like they did.”) A good story, as the poet Samuel Taylor Coleridge wrote, automatically triggers a “willing suspension of disbelief for the moment.”

"The South Sea bubble brimmed with “story stocks.”"

Like companies could "develop an “air pump for the brain,” convert sewage into gunpowder"

"The South Sea Co. had its own story: Originally created to sell enslaved people to Spain’s American colonies, by 1720 it had morphed into a complex scheme for refinancing the British government’s massive war debts."

Tuesday, July 28, 2020

How the White-Black median household income gap has changed over time and how that compares to the Asian-White gap

The post from two days ago covered the Asian-White gap, which has generally been growing since 1988 (the first year the Census Bureau reports Asian incomes). Since 1988, Asian income has always been higher than White income.

One thing I will show later is that the Asian-White gap as a percentage of the White-Black gap has been growing.

The graph below shows the White-Black gap since 1967 (the first year the Census Bureau reports median household income). Everything is adjusted for inflation (in 2018 dollars). I used the BLACK ALONE/BLACK and the WHITE ALONE/WHITE series (see the earlier post for an explanation).




The gray line shows the gap. It has been growing in recent years. It was about $25,000 in 2018, close to what it was in 1988. The table below has all the numbers.


Year
White
Black
Diff
1967
49,102
28,510
20,592
1968
51,138
30,155
20,983
1969
53,163
32,134
21,029
1970
52,646
32,044
20,602
1971
52,354
30,926
21,428
1972
54,759
31,963
22,796
1973
55,809
32,851
22,958
1974
53,927
32,071
21,856
1975
52,512
31,524
20,988
1976
53,474
31,797
21,677
1977
54,020
31,878
22,142
1978
55,470
33,335
22,135
1979
55,839
32,784
23,055
1980
54,361
31,318
23,043
1981
53,577
30,065
23,512
1982
52,943
30,005
22,938
1983
52,662
29,885
22,777
1984
54,586
31,096
23,490
1985
55,588
33,072
22,516
1986
57,411
33,076
24,335
1987
58,222
33,231
24,991
1988
58,900
33,577
25,323
1989
59,619
35,456
24,163
1990
58,359
34,898
23,461
1991
56,920
33,909
23,011
1992
56,665
32,995
23,670
1993
56,560
33,519
23,041
1994
57,198
35,344
21,854
1995
58,705
36,755
21,950
1996
59,413
37,543
21,870
1997
60,990
39,202
21,788
1998
63,170
39,143
24,027
1999
63,989
42,196
21,793
2000
64,216
43,380
20,836
2001
63,293
41,899
21,394
2002
63,107
40,628
22,479
2003
62,451
40,573
21,878
2004
62,177
40,105
22,072
2005
62,584
39,774
22,810
2006
63,264
39,913
23,351
2007
63,270
41,176
22,094
2008
61,160
40,006
21,154
2009
60,845
38,228
22,617
2010
59,682
37,077
22,605
2011
58,423
36,061
22,362
2012
58,846
36,510
22,336
2013
61,268
38,140
23,128
2014
60,376
37,583
22,793
2015
63,710
39,108
24,602
2016
64,729
41,323
23,406
2017
66,864
41,239
25,625
2018
66,943
41,361
25,582


The next chart does ratio instead of absolute difference. Before 1990, Black income was almost always 60% or less of White income while since then it is almost always been above 60%. But this looks like slow progress.



This next chart shows the Asian-White gap as a percentage of the White-Black gap.  The long term trend is for it to grow.

In 1988, White income was 58,900 while it was 66,034 for Asians. That is a gap of 7,134. From the table above, the White-Black gap was 25,323. So the ratio was 28%.

The Asian-White gap in 2018 was 20,251 in 2018. The White-Black gap was 25,582. That is a ratio of about 79%.