4 ms·

Well, yes, but you have to draw the line somewhere. A long-term diet of ramen is not healthy, but drawing up a reference basket of middle-of-the-road nutritious

by Sanzig 6y ago

Well, yes, but you have to draw the line somewhere. A long-term diet of ramen is not healthy, but drawing up a reference basket of middle-of-the-road nutritious food and pricing it based on median market prices for statistical purposes isn't unreasonable. And we can agree that assuming that somebody uses the library for the internet is unreasonable - a better baseline is a basic package for both at local market rates. The goal is to assess reasonable flexibility in cashflow. If the study author is biased towards trying to prove a predetermined outcome (eg: that minimum wage in high CoL areas is livable), the results are poisoned from the get-go: good luck getting repeat business from McDonalds' if your study unearths some unpleasant truths. It's the classic case of a metric becoming a goal and ceasing to be useful as a metric. I do think this is certainly something a national statistical agency could do in a minimally biased fashion, if done rigorously.

AstralStorm 6y ago
Then again, not having resources or entertainment while having a soul-crushing boring job is a massive mental health risk. Ignoring that is of course a nasty bias. Much like ignoring that you do sometimes have to buy clothing or shoes, you probably do have to spend on something to unwind. (What it is depends on the person.) Likewise, ignoring cost of transportation is a chief sin of those analyses, even bus tickets are not free, and bicycles are not exactly fast.
lumost 6y ago
The problem with average and basket goods for this problem is that it's difficult to assess whether someone is financially constrained when their circumstances push them beyond the average. As a minimal example, calorie and nutrition demands can vary by about 2x with height and biological factors and easily up to 3-4x if you consider more active lifestyles. You would generally expect some portion of the population in the study to need double or even quadruple the grocery cost per household member than the average. Are they cost burdened if most of their budget goes to grocery? Similarly housing is heavily biased by when you purchased property comparing someone renting market rate housing to someone who lives in rent control or has owned property for a while is challenging. You can't exclude owners as they are the most likely to bankrupt in this scenario if they become cost burdened. Even worse, for both of the above baselines the cost of goods can easily vary by 4x depending on where you live within a 50 mile radius when it comes to dense population centers. Starbucks employees in SF proper aren't going to commute 4+ hours through bay traffic to live in lower cost regions of California. They will inherently need to purchase grocery, housing, and other goods at higher than average prices. These edge cases matter in this analysis as the outliers are the ones who probably will go bankrupt or otherwise exit the economy if the economic environment turns profoundly negative. If by circumstance the average household is happy, but 10% of households are on the brink - it wouldn't take much for things to take a serious turn. All that being said, there is no reason a government agency couldn't devise a methodology based on the 90th or 99th percentile to capture these edge cases and report on a number that would more reasonably reflect the long-term outlook of houses under financial stress.

by Sanzig 6y ago

AstralStorm 6y ago

Then again, not having resources or entertainment while having a soul-crushing boring job is a massive mental health risk. Ignoring that is of course a nasty bias. Much like ignoring that you do sometimes have to buy clothing or shoes, you probably do have to spend on something to unwind. (What it is depends on the person.) Likewise, ignoring cost of transportation is a chief sin of those analyses, even bus tickets are not free, and bicycles are not exactly fast.

lumost 6y ago

The problem with average and basket goods for this problem is that it's difficult to assess whether someone is financially constrained when their circumstances push them beyond the average. As a minimal example, calorie and nutrition demands can vary by about 2x with height and biological factors and easily up to 3-4x if you consider more active lifestyles. You would generally expect some portion of the population in the study to need double or even quadruple the grocery cost per household member than the average. Are they cost burdened if most of their budget goes to grocery? Similarly housing is heavily biased by when you purchased property comparing someone renting market rate housing to someone who lives in rent control or has owned property for a while is challenging. You can't exclude owners as they are the most likely to bankrupt in this scenario if they become cost burdened. Even worse, for both of the above baselines the cost of goods can easily vary by 4x depending on where you live within a 50 mile radius when it comes to dense population centers. Starbucks employees in SF proper aren't going to commute 4+ hours through bay traffic to live in lower cost regions of California. They will inherently need to purchase grocery, housing, and other goods at higher than average prices. These edge cases matter in this analysis as the outliers are the ones who probably will go bankrupt or otherwise exit the economy if the economic environment turns profoundly negative. If by circumstance the average household is happy, but 10% of households are on the brink - it wouldn't take much for things to take a serious turn. All that being said, there is no reason a government agency couldn't devise a methodology based on the 90th or 99th percentile to capture these edge cases and report on a number that would more reasonably reflect the long-term outlook of houses under financial stress.