George Rebane

The exodus of the producers and the productive from California has been a recurring theme on RR ever since the Democrats assumed total control, turned the state into the nation’s home for welfare recipients, started raising taxes and fees as if there were no tomorrow, and opened its gates to illegal aliens.  Essentially, the Democratic super-majority has turned the state into a sanctuary from sanity.  Among the reams of data presented to buttress the exodus argument, a reader sent me the graphic below that highlights just one aspect of nationwide migrations – and these are NET migrations – from/to our country’s prominent metropolitan areas (more here).

This shows that from LA alone we have a net outflow of 46,720 people per year.  And do you think these are the gimmes who think they’ll get more sanctuary protections and better government bennies from some other state?

The point I want to make here is how Sacramento’s progressive elites continue convincing their grass roots constituencies of the idea that California is not losing people who build, work, and are employed by its businesses.  Granted, as the Dept of Education has told us for years, CA produces, acretes, and aggregates some of the dimmest bulbs on the national intellectual scale.  And Sacramento Democrats have the whole-hearted support of the state’s lamestream media to promote the propaganda bilge that gets pumped out of the capitol.  But what makes that stream of fake news so effective is not that the progressive elites are ignorant, they have the same information RR readers have and perhaps more.  Instead these lying leftwingers range somewhere between the terminally cynical and evil in justifying why they dispense their drivel on demographics.  They know that their constituents can’t tell the difference between the state’s population growth – driven mainly by the influx of the indigent and lower quintile minorities (including illegal aliens) – and the exodus of middle-class workers, upper quintile taxpayers (including the dispersal of big corporations’ facilities, and companies with major markets outside CA).

The evil part comes from constantly taking advantage of their grassroots true believers whom they know from national stats to range between stupid and ignorant.  (Recall, ‘stupid’ means lack of processing power, you’re more or less born with stupid; ‘ignorant’ denotes having a poor knowledge base, which for the non-stupid can be remedied with education and reading.)

A seminal characteristic of the stupid is that their beliefs are more than less immune to new information that counters what is already firmly entrenched.  They don’t embrace Lord Keynes’ refreshing dictum, ‘When new information comes in, I change my mind.  What do you do?’  They often reveal themselves in conversational settings with the preamble, ‘There’s nothing you can say that …’.  In the fields of inference and decision theory, their anchoring beliefs are technically represented with Bayesian probability priors that equal either unity (firm belief that a proposition is TRUE) or zero (firm belief that a proposition is FALSE).  With such starting points, no amount of new evidence is able to budge their prior beliefs, i.e. reduce or increase the belief in a proposition when new evidence arrives.

In our current example of the California Exodus, no data from reliable sources will change the minds of the state’s stupid.  If they have been told to believe that there is no exodus, then there it shall remain – the case is settled.  One can illustrate similar behavior with, say, ‘climate change’ and a number of other publicly bandied and politically momentous propositions.

And in recent years with the influx of coastal collectives (Nevada County is turning blue), our own county now serves as a posterchild of the above-described phenomenon, especially as it comes to CA emigration.  Examples of local leftwing loonies abound both in the regional media and the blogosphere.  I mean there is some serious stupidity afoot around here.

ADDENDUM – [Trigger warning – the following is recommended only for the curious and algebraically astute who want to see and, perhaps, even apply the quantitative aspects of Bayesian learning.  BTW, this formulation also forms the basis of how deep learning networks update themselves with new data.]

Here on RR we have taken journeys through Bayes theorem and Bayesian inference several times over the years, so we won’t have to start from the beginning.  I refer you to ‘Bayes sans equations and tears’ as one my more illuminating dissertations on this for the layman.  There the likelihood form of Bayes is illustrated asTo see how the firmly entrenched (‘There’s nothing you can say …’) stupid operate in this context, let’s take the hypothesis H = ‘There is no CA exodus’.  As evinced by some local leftwing true believers, their probability P(H) = 1, i.e. it is a certainty that there’s no CA exodus as described.  So now some evidence E comes in from, say, the Dept of Commerce.  And E = numerical dataset showing net exodus of the productive and producers.  Now the decision maker has to form the likelihood ratio L(E|H) = P(E|H)/P(E|¬H) of the two conditional probabilities – P(E|H) is the probability of such evidence occurring when the hypothesis is true, and P(E|¬H) is the probability of such evidence occurring when the hypothesis is false.  In our case P(E|H) is that DepComm issues erroneous data in the face of no exodus, and P(E|¬H) is that DepComm issues such data correctly when the hypothesis is false, i.e. there is in fact such an exodus.

Even the nation’s double dummies don’t think it’s very likely that DepComm will issue such grossly erroneous data.  Let’s assume they think that at most P(E|H) = 0.01 or one chance in a hundred that the data is wrong.  Conversely, given their past record of putting out good data on US population dynamics, they assess that P(E|¬H) = 0.999, i.e. they almost certainly got it right.  Then our likelihood ratio from above is L = 0.01/0.999 = 0.01.  If we insert these into the Bayes equation to find out the impact of new evidence on the double dummy, we have his updated belief that there is no CA exodus as the probability P(H|E) that there is no exodus (H) given the DepComm data (E)

P(H|E) = (0.01*1)/(0.01*1 + 1 – 1) = 1

In short, P(H|E) = P(H) = 1, which means that the evidence from DepComm, no matter how reliable, made absolutely no impact on him.  Putting in P(H) = 0, the other extreme belief, shows again that P(H|E) = P(H) = 0, no matter what the reliability of the newly introduced evidence.  This gives the formal proof that such people with firmly anchored beliefs are beyond the reach of reason, no matter how convincing such evidence is to the rest of us.

We’ll end with a quantitative example of someone with an open mind who at the start has no idea whether there is or is not such an exodus.  Quantitatively this belief (i.e. ignorance of the hypothesis being TRUE or FALSE, is represented by the Bayesian prior P(H) = 0.5, i.e. it can be true or false on a 50-50 basis.)  When such a more reasonable person is given the DepComm evidence, he uses the Bayes formula to update his belief as follows.

P(H|E) = (0.01*0.5)/(0.01*0.5 + 1 – 0.5) = 0.0099 ~ 0.01

This says that the DepComm evidence reduced his belief in the ‘no exodus’ hypothesis from 0.5 to 0.01 or about 50-fold.  Then with this new evidence this individual’s belief in the existence of a CA exodus is updated to 1 – 0.01 = 0.99, or almost certainty (but not quite).  So there you have the technical version of the difference between an educable person and one terminally stupid.  With prior beliefs anchored in bookends of certainty (0 or 1), there literally is nothing you can say or no evidence you can present that will change their minds.