3 ms·
>...social engineering... Like joint tax returns, mortgage deductions, tax exemption for churches, child tax credits, and narrow definitions of marriage?
by TomMckenny 8y ago
>...social engineering...
Like joint tax returns, mortgage deductions, tax exemption for churches, child tax credits, and narrow definitions of marriage?
- gizmo686 8y agoI would say like a narrow definition of marriage [0]. However, the prejoritive I attempted to convey was not actually that it is social engineering, but rather that it is a "crude and ill considered attempt [at social engineering]". Notice that all of your other examples are taxes, which are a well established and effective mechanism for social engineering. I would probably still disagree with it, but I would not be calling the policy stupid if it was a subsidy for women in boards of directors, instead of being an unfunded mandate. [0] Actually, I would say the entire legal institution of marriage is, but I digress.
- TomMckenny 8y agoPerhaps not your intent, but the labels "social engineering" vs "traditional values" are so widely used by pundits to replace inquiry with highly emotional obedience that they have become suspect. For incentive vs mandate, much of the country has restrictions on alcohol sales. Often just on Sunday, obviously an attempt to encourage a particular religious practice. Some limit sales to government run companies. These are not incentive systems but direct prohibition. Yet there is no huge outcry of "social engineering" or "socialism" from those who condemn other policies with those terms, instead theses practices are frequently defended as "traditional values" or "community values". For California, an incentive system might have been better received or perhaps more efficient, if slower, toward reform. But in practice, a tax cut is a tax increase to those who don't get it but must pick up the revenue slack. For example, if we look at it honestly, the mortgage deduction is little more than a transfer of wealth from the poor to the middle class. It's worth noting the study cited, taken in narrow time window, reporting effects on the level of 1.5% during a period of large market fluctuations, should probably be taken with a grain of salt. And even if accurate, could be due to things such as uncertainty caused by any shuffle of board members regardless of gender. But even if their cause is correct, the California public seems willing to endure a small short term loss to share holders if the result is a disruption to discrimination networks both known and unknown. It is also reasonable to believe that when irrelevant barriers to promotion (including gender) are reduced, over all efficiency will increase with the larger pool of candidates. And that this will take more than three months. Judging from California's overall prosperity, their often criticized techniques do seem to work well even if measured just solely on wealth.
- solidsnack9000 8y agoSome of these might not have started as legislative attempts at social engineering at all, if they reflected the situation prior to the state legislating in those areas. For example, tax exemptions for churches pre-date every government in existence today.