Monday, July 17, 2017

Q&A - 17/7

Economist

One of the economy’s biggest mysteries is this: The labor market is the strongest it has been in a decade, yet wages are rising barely faster than inflation. For some reason, the booming job market and ultralow unemployment rate, which fell to 4.4 percent in April, haven’t led employers to raise pay in a meaningful way. That flies in the face of a basic assumption of how the economy works: A tight labor market is expected to lead to pay increases [..]

No mystery

This person is using the wrong unemployment number, his model tells him for that incorrect unemployment there has to be more wage growth, that's why he is mystified. If we use the real unemployment, i.e. the number that is opposite of employment, the model's fine. We take prime age labor participation rate and literally subtract that from 100%, that's unemployment. Then with help from this post, and using data from FRED, we fit a linear model between unemployment and wage growth,

import pandas as pd, datetime
from pandas_datareader import data
import statsmodels.formula.api as smf

start=datetime.datetime(1970, 1, 1)
end=datetime.datetime(2017, 1, 1)
cols = ['ECIWAG','LNS12300060','UNRATE']
df = data.DataReader(cols, 'fred', start, end)
df = df.dropna()
df['ECIWAG2'] = df.shift(4).ECIWAG
df['wagegrowth'] = (df.ECIWAG-df.ECIWAG2) / df.ECIWAG2 * 100.
df['unemp_real'] = 100. - df.LNS12300060

results = smf.ols('wagegrowth ~ unemp_real', data=df).fit()
print results.rsquared_adj, results.params.unemp_real
plt.scatter(df.unemp_real, df.wagegrowth)
plt.plot(df.tail(1).unemp_real, df.tail(1).wagegrowth, 'gd')
xx = np.linspace(20,25,100)
yy = results.params.Intercept+results.params.unemp_real*xx
plt.plot(xx,yy)

The fit is good. [geek] For the naive model using UNRATE the fit reports R^2 0.61, the model above reports 0.81 which is clearly superior. [/geek].

How about the larger question, is annual wage growth of 2.5% any good? Maybe. Model says in order to get to annual 4.0%, (real) unemployment needs to fall by 5% which is huge. Graph,



Walter Isaacson

[Einstein intervenes in a public protest at Reichstag, 1918] Years later, when Adolf Hitler and his Nazis were in power, Einstein would ruefully look back on that day in Berlin. “Do you still remember the occasion some 25 years ago when we went together to the Reichstag building, convinced that we could turn the people there into honest democrats?” he wrote Born. “How naïve we were for men of forty.”

Hah

Larger forces play themselves out. It's not possible to change history with a few great speeches, symbolic moves even if those speeches were given by Einstein. Not that the small cannot change things -they can if they are part of a larger dynamic :) so act on them- but the overall result might not always turn out exactly how everyone wants.

And usually the largest force in play is geography.

G. Friedman

[From The Next Decade] History had placed Germany on the north of the North European Plain, an area with a few rivers to serve as defenses, but some of the most productive parts of this new nation-state were on the opposite bank of the Rhine, completely unprotected. To the west was France. To the east was Russia. Both had enjoyed the centuries when Germany was fragmented and weak, but now there was a frightening new Germany, economically the most dynamic country in Europe, with a powerful military and with a deep sense of insecurity.

Germany in turn was frightened by its neighbors’ fears. Germany’s leaders knew their nation could not survive if it was attacked simultaneously by France and Russia. They also believed that at some point such an attack would come, because they understood how intimidating they appeared to their neighbors. Germany could not permit France and Russia to start a war at the time or place of their choosing, and thus Germany, driven by its own fear, devised a strategy of preemption coupled with alliances.

Europe in the twentieth century was defined by these fears, which, being imposed by geography, were both rational and unavoidable [..]

Both world wars were launched according to a single scenario: Germany, insecure because of its geographical position, swept across France in a lightning attack. The goal in both cases was to defeat France quickly, then deal with Russia. In 1914, the Germans failed to defeat France quickly, the troops dug in, and the conflict became a protracted war. The Germans found themselves fighting France, Britain, and Russia simultaneously in both the east and the west.

Interesting

Tim Harford, FT

If we are to take the best advantage of a true gig economy, we need to prepare for more radical change. Governments have been content to use corporations as delivery mechanisms for benefits that include pensions, parental leave, sick leave, holidays and sometimes healthcare — not to mention the minimum wage. [..]

We will need governments to provide essential benefits, perhaps minimalist, perhaps generous, to all citizens. Above that safety net, we need portable benefits — mentioned warmly but briefly by Mr Taylor — so that even a 10-minute gig helps to fill a pension pot or earn time towards a holiday. Traditional corporate jobs have been socially useful, but if you push any model too far from reality, it will snap.

Yes

Let me add, even before the spread of the gig economy everywhere (today's situation), the existing system already started to show signs of major fail, so it is in our interest to make the switch sooner rather than later.

Anu Partanen

[From The Nordic Theory of Everything] Among the less privileged, meanwhile, American marriage appeared to be in a state of full-blown crisis. A much-discussed study had found that among white people in their thirties and forties, less than half of those with only a high-school education were tying the knot. Critics debated the causes of this reality, but what was remarkable to me was that this discussion, too, ultimately circled around the financial axis [..] In America, if you were contemplating getting married and starting a family, you first needed to think very carefully about your finances. How much debt did you have in student loans? Could you afford health insurance? For that matter, how much would it cost just to give birth? The maternity benefits of different health-insurance plans varied dramatically. I was stunned when I learned of a young couple who had health insurance, and nevertheless had ended up owing the hospital twenty thousand dollars for the birth of their baby. [..]

One of the most heartbreaking episodes I witnessed after moving to the United States involved an American acquaintance who was battling cancer. To make matters even worse, it was clear that the person’s domestic relationship was fraying at the same time. The peculiarly American twist on the story was this: If the couple split up, the young cancer patient, with many months of expensive treatment ahead, would be left without health insurance, since it was being provided through the domestic partner’s employer. The unhappy relationship seemed to last much longer than it should have, hurting everyone involved far more than necessary. The trauma of the situation had been vastly multiplied simply because of the total dependency of everyone in the equation on the employer. [..]

Far less tragic than the story of my friend with cancer [..] were the cases of people taking a job they didn’t really want, simply because they needed the health insurance. Others hesitated to change jobs, or decided not to make an otherwise positive career move, because they’d have to give up their health coverage. Less obvious than the matter of health care, but also insidious, was the hesitancy of practically every American I met to take their full allotment of vacation time, as allowed by their employer, no matter how paltry. Never mind actually leaving work every day at five.

Gradually it dawned on me how much people in America depended on their employers for all sorts of things that were unimaginable to me: medical care, health savings accounts, and pension contributions, to name the most obvious. The result was that employers ended up having far more power in the relationship than the employee. In America jeopardizing your relationship with your employer carried personal risks that extend far beyond the workplace [..]

Americans have a reputation for changing jobs often, but as a result of their dependence on employer-provided benefits, my American acquaintances all seem far more beholden to their bosses than the Nordics I know. Americans hardly take any parental leave, and clearly feel obligated to work extremely long hours, with little say over how these hours are arranged. By contrast, I had worked a number of high-pressure jobs as a journalist in Finland [..] it wasn’t even in the realm of possibility that all my health care could be compromised by anything to do with my job.

By now I was used to hearing the Nordic countries dismissed as “socialist nanny states.” But ironically it was here in America that businesses trying to manufacture products and make a buck had somehow gotten saddled with the nanny’s job of taking care of their employees’ health. Surely, I thought, Milton Friedman, the great free-market economist, must be turning in his grave! From a Nordic perspective, it seemed ludicrous to burden for-profit companies with the responsibility of providing employees with such a fundamental, complicated, and expensive social service.

People in the United States were aware of this contradiction, of course, and in discussions of the American business landscape, experts often pointed to the burdens that health-care obligations placed on companies, especially on small businesses. But no one seemed to be talking about the other side of the coin: the unhealthy dependence on employers that this creates among employees receiving, or hoping to receive, these benefits. It was an old-fashioned and oppressive sort of dependence, it seemed to me, completely at odds with the modern era of individual liberty and opportunity.

The fast-paced, stressful nature of modern life in a globalized world might be inevitable, but leaving people to muddle through it by falling back on old-fashioned family- and village-based support structures that no longer functioned the way they once did—that was by no means inevitable. The more I experienced life in the United States, the more I began to think that what Nordic societies [..] had succeeded in pushing past outdated forms of social dependency, and that they had taken modernity to its logical conclusion.

Nice

Question

What is so modern about government provided healthcare?

Everything about it

Gov-funded healthcare programs, with their centralized, concentrated, synchronized slant, are truly in the realm of second wave. A third-wave, post-modern system could have a person's mobile phone to sell personal "life bonds" on a blockchain the moment person gets sick, the bonds would have maturity of, say, 20+ years (buyers weigh his probability of staying alive), system would offer high return so it is attractive to buyers and gets him the money he needs for care while the system simultaneously short-sells the care provider company stocks the patient did not choose, at the same time electronically bartering Starbucks coffee cash + social net like money for care money. Do little bi'ness, get little Remy, little Henny, little Crissy, know wha Im sayin'? But now we are in the realm of scifi - we simply do not have all of this yet.

Isaac Asimov

The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' but 'That's funny..."

The Great Asimov

Documentary 

The compass was the greatest invention for naval navigation.

It was not

This documentary tries to paint China as some big-time discoverer nation (compass was discovered there). If China was so great at the seas, then why did they not discover / populate Australia? It's right under their nose. Right? Wouldn't it be great having that major piece of land instead of getting stuck with a billion people on your current land?

Large-scale oceanic navigation requires a global measure. At the very least the sun for the latitude through something like the sextant, and a clock for longitude. Clock based navigation did not take off until the beginning of Renaissance.

Dave Sobel

[From Longitude] [T]he sea captains of [old] relied on “dead reckoning” to gauge their distance east or west of home port. The captain would throw a log overboard and observe how quickly the ship receded from this temporary guidepost. He noted the crude speedometer reading in his ship’s logbook, along with the direction of travel, which he took from the stars or a compass, and the length of time on a particular course, counted with a sandglass or a pocket watch. Factoring in the effects of ocean currents, fickle winds, and errors in judgment, he then determined his longitude. He routinely missed his mark, of course—searching in vain for the island where he had hoped to find fresh water, or even the continent that was his destination. Too often, the technique of dead reckoning marked him for a dead man.

Exactly

Democracy, Bitcoin

A summary of comments made on democracy and bitcoin .