trends

Some time ago Linkedin started sending weekly e-mail with the trends for "your kind of jobs". Every week since it's been reporting a 10% drop in job postings, then went over 10%. Well, last week it held at 0, then this week it dropped another 13%. 

At the same time the government reports tout the job growth. So, is it a drop in high-paying jobs and growth in low-paying jobs? (Which is not exactly a great thing). As it happens, I know someone who runs a job posting web site that is used predominantly for low-to-mid paying jobs. And guess what, he says that the postings on his site have also sharply dropped around April. So no, not a growth of low-paying jobs either.

 What does it all tells us about the government statistics? As they saying goes, "lies, damned lies, and statistics". Maybe they'll "revise" it a few months later as they've done with statistics on income after inflation, where the loudly touted slight growth above inflation had quietly turned into a ~5% loss.

networking & financials

Looking at the Federal Reserve rates

https://www.macrotrends.net/2015/fed-funds-rate-historical-chart ,

we can observe that once in a while they suddenly decide to raise the rates quickly, each time followed by a recession. As the people who caused the Great Depression said,  they "fought to stop the speculations", and succeeded, bringing the whole economy down with them.

But the graph overall looks very reminiscent to what we see in the congestion control protocols. Except that the congestion control protocols act more sensibly: increase the transmission speed slowly, but when a catastrophe is detected, drop it sharply. It would be interesting to build an economic model and use the congestion control protocols to regulate the refinancing rate in it, starting with the classic TCP. I suspect that it might be able to do a lot better than Fed's current tealeaf reading, since the Fed seems to be doing things backwards: raising the rates sharply and dropping slowly. Since the catastrophes happen when the rates get too high, it would make a lot more sense to limit the rate of increases to something like 0.25% (25bpp) per year, to creep up on that catastrophe slowly, detect it early in the making, and then react sharply.