Yeah it really is inefficient to appease the middle income voters compared to the poor or the wealthy ones. I do mostly agree to what was mentioned by Cliff but I’ll just add some points which have been mentioned in the original post as ‘to be written later.’
I see the bump-up of the poor and unhappy middle class as a nice dilemma. I like such changes of political landscapes, brought by increase of the disposable income thanks to social security policies. Though discrepancies between the targets of income policies and their approval responses can bring weird situations such as the poor voters getting angry when it won’t have any significant impact on the low income, I think it’s kinda unavoidable unless the gap between original v disposable income is added as an approval factor of individual voters. (and it can also be problematic as tax policies will have less nominal approval penalties) But still, I have some points to mention regarding this matter.
First of all, it is way too easy to eradicate the poverty. It doesn’t take that much to reduce the poverty stat down to zero, and the same goes for the poor group membership. Not to mention that there are a number of highly efficient anti-poverty or low-income boosting policies, even ones that aren’t targeting the poor can significantly boost incomes of large demographics (such as ones increasing state employees income & membership), making it possible for the poor voters to get into the middle income group quite easily. This may have been caused by either excessively high income boosts or way too small disposable income loss from taxation. However, introducing harsher disposable income loss to the income tax could cause poor-middle income reversal of disposable income distribution as low income benefits & middle income taxation will be applied to the original incomes.
Second, poverty eradication generally relies on residual welfare programs which is overpowered. Such welfare policies neither make the middle class dissatisfied nor have downsides such as administrative burdens of recipient selection process, boundary problems, or a failure to reach those in need. And there’s more. A nice example would be the Food Stamp policy. Its cost is linked with the poor voter membership but its anti-poverty effect and low income boost don’t disappear just because there’s no one to receive food stamps and no budget to execute. I do see that there’s little reason to address such matter, as this game effectively lacks economic recession or bust and thus such welfare programs will only be reactivated when the player deliberately decides to boost poverty. Nonetheless, I think it would be nice if such measures can have their limitations or downsides. Though I fear that it might cause oscillation of poor-middle income membership, I think those selective welfare policies need to lose their effects when people starts to get out of poverty. They should be seen as cost-efficient and temporary measures, not something gives permanent benefits.
Lastly, there are some sub-groups who aren’t likely to make their living without state provision but I can safely cut public spending. For example, at least some fraction of the retired must have been poor when they were younger and thus couldn’t afford private pension. These poor-retired people have little chance of living without state pension / state healthcare / social care but it seems they can. Because A) there’s no sub-division on the retired, B) retired people in general don’t have innately low original income, C) private healthcare doesn’t have any negative impact on retired income, and D) scrapping state pension will only replace state-to-retired transfer of income with middle income-to-retired (by private pension), causing decrease in neither retired income nor low income.
Another example would be the “invisible” unemployed. It’s quite easy to have a massive unemployment while playing but it seems there’s no income loss for having high unemployment. In fact, it’s the opposite - the poor will gain income from the unemployment benefits, which has low income boost proportional to the unemployment stat. Cutting the unemployment benefits will lead to a slight fall in the poor income but I still don’t need to address high unemployment or reconsider cutting benefits.
Addressing these matters might need way too much works but, in my opinion, there are some low-hanging fruits to start with. While adding poor_perc to food stamp effects as a second factor might need more experimentation, some adjustments will be the first step of addressing the #1. Also, if you are willing to just lump sub-group poverty together with its broader group, things might become simpler. Just adding poverty-to-retired income link could roughly work I guess. Or maybe add private healthcare-to-retired income link. Regarding unemployment, I guess wage-to-income effects also can have unemployment as a second factor.
I think I’ve written way too long ;D Thank you for reading a wall of text.
tl;dr
- disposable income boosts are way too strong in general (or income losses from taxation are way too weak)
- food stamp-like policies are broken and it’s not just a matter of adjusting numbers
- retired without state pension or unemployed without benefits don’t cause any poverty