I enjoyed the article, but your calculation for workers' benefit at the end is vary wrong.
a. -0.23 is not the effected workers elasticity, it usually includes majority of people earning more then the new MW. If only 33% of them are effected, then only 33% had an increase in wages.
b. The increase is not 10% for all effected workers, because many earn something between the old MW and the new MW, so for them the increase might only be 5% for example. You should use about 6-7% increase in effected wages for 10% MW increase.
a+b mean you should use 6-7% increase for the 33% effected workers, which would give you:
7%*(0.33-0.023)-100%*0.023 = -1.6%
an immediate decrease in wages even for 10% MW increase.
c. Money is not everything and at the end they could and probably would suffer from worse job conditions.
Why do you need wage data to calculate elasticity?
let's say hypothetically that the true effect is that no one's wage increased, but what happens is that all the people that earned less then the new MW are fired.
In that case the wages won't change. So why do they even matter? the average wage would change but that is useless information.
Using change in wages as the denominator instead of change in the minimum wage has two benefits: 1) it makes it possible to include MW increases from zero (i.e., a country enacting a minimum wage for the first time; if I used the change in MW as the denominator, it would be +infinite %); and 2) it puts all MW increases on a common scale. Obviously an increase in the MW from $8 to $10 an hour in a rich country would affect way more people than raising it from $4 to $5 in some other rich country, and so would bring up wages by a higher % than the latter increase, despite them both being the same MW % change.
As for your other comment, the design of these studies is supposed to provide only causal results. Obviously they're not perfect methodologically, and chance variation still affects them. But, in a large pooled analysis, the average effect should probably work out such that estimated wage change is roughly equal to the causal change, and so it doesn't matter if you're including unaffected workers; either those workers would have a wage change of zero, or they'd be affected indirectly (a spillover effect) and so their wage change would still be causally related to the MW.
I agree with your point about employment being worth more than what you'd estimate based on wages, especially since low wage labor is usually work you get when first enter the labor market and are trying to move your way up. However, it is difficult to assign any specific number to the economic affect of job loss apart from wage loss; perhaps I'll be able to do this sometime in the future.
Ok now I understand that you used OWE instead of MWE, but because at the end both are pretty similar in magnitude, there shouldn't be a big difference in the average wage change. But as I showed when using OWE you vary easily get a negative change, so how could this be?
So I suggest that maybe if the most low-paying jobs are lost, then there could theoretically be an "increase" in wages without any real increase. So at least some of the "increase" measured in the OWE denominator isn't real, and when properly calculated (total wage change instead of average) the OWE would be higher.
The elasticity the entire time is OWE, though I can see how you could get confused, since I didn't explain the difference between MWE and OWE until near the end. So you don't need to make any adjustments to the calculations in section 5; they're already in terms of OWE
I understand, but according to my calculation using MWE I got -1.6% total wage change, so now I'm trying to understand why there's such a huge difference when using MWE vs OWE - this shouldn't be the case!
So I think the problem is that you assume that a 10% increase in avg wage means that you can calculate the increase as 98%*10%, when in reality if the increase in avg wage is due to low wage workers getting fired, then this is incorrect, because theoretically the increase can also be 98%*0%.
let me copy my other comment here so people understand what I'm talking about:
"a. -0.23 is not the effected workers elasticity, it usually includes majority of people earning more then the new MW. If only 33% of them are effected, then only 33% had an increase in wages.
b. The increase is not 10% for all effected workers, because many earn something between the old MW and the new MW, so for them the increase might only be 5% for example. You should use about 6-7% increase in effected wages for 10% MW increase.
a+b mean you should use 6-7% increase for the 33% effected workers, which would give you:
7%*(0.33-0.023)-100%*0.023 = -1.6%
an immediate decrease in wages even for 10% MW increase.""
The issue is that the elasticity of -0.23 already is an OWE. The MWE is a completely different number which I did not calculate. The OWE already accounts for the fact that not everybody is affected; if someone's wages don't rise as a result of the MW change, then they won't affect the wage change which is the denominator for MWE
From what I saw the number for MWE it should be vary close to the OWE. The difference isn't big enough to explain the problem.
I'm not sure about that, the calculation of the average change is usually not done on an individual basis, are you sure that this is how the calculation is done in practice?
1- the minimum wage the higher it is, the bigger the burden for businesses, employees and employ-seekers this last 2 via the employment costs for the first one
2- big businesses are more favoured by MW than smaller ones as long as the formers have a higher financial effort and could be more easily displaced or prevented from entry to the market, this logic also applies to skilled and/or experienced workers in comparison to those who are less although here is more variable because some arguably lesser titulations, based in academical-education system and titulation hierarchy consensus, have a higher chance of employment than supposedly higher ones
2.1- Usually youth have less experience than older people
3- MW positive correlations relating to higher employment doesn't really exist is only that thanks to other situations and policies they constraint it's bad effects basically we succed despite of not thanks to this
I enjoyed the article, but your calculation for workers' benefit at the end is vary wrong.
a. -0.23 is not the effected workers elasticity, it usually includes majority of people earning more then the new MW. If only 33% of them are effected, then only 33% had an increase in wages.
b. The increase is not 10% for all effected workers, because many earn something between the old MW and the new MW, so for them the increase might only be 5% for example. You should use about 6-7% increase in effected wages for 10% MW increase.
a+b mean you should use 6-7% increase for the 33% effected workers, which would give you:
7%*(0.33-0.023)-100%*0.023 = -1.6%
an immediate decrease in wages even for 10% MW increase.
c. Money is not everything and at the end they could and probably would suffer from worse job conditions.
Why do you need wage data to calculate elasticity?
let's say hypothetically that the true effect is that no one's wage increased, but what happens is that all the people that earned less then the new MW are fired.
In that case the wages won't change. So why do they even matter? the average wage would change but that is useless information.
Using change in wages as the denominator instead of change in the minimum wage has two benefits: 1) it makes it possible to include MW increases from zero (i.e., a country enacting a minimum wage for the first time; if I used the change in MW as the denominator, it would be +infinite %); and 2) it puts all MW increases on a common scale. Obviously an increase in the MW from $8 to $10 an hour in a rich country would affect way more people than raising it from $4 to $5 in some other rich country, and so would bring up wages by a higher % than the latter increase, despite them both being the same MW % change.
As for your other comment, the design of these studies is supposed to provide only causal results. Obviously they're not perfect methodologically, and chance variation still affects them. But, in a large pooled analysis, the average effect should probably work out such that estimated wage change is roughly equal to the causal change, and so it doesn't matter if you're including unaffected workers; either those workers would have a wage change of zero, or they'd be affected indirectly (a spillover effect) and so their wage change would still be causally related to the MW.
I agree with your point about employment being worth more than what you'd estimate based on wages, especially since low wage labor is usually work you get when first enter the labor market and are trying to move your way up. However, it is difficult to assign any specific number to the economic affect of job loss apart from wage loss; perhaps I'll be able to do this sometime in the future.
Ok now I understand that you used OWE instead of MWE, but because at the end both are pretty similar in magnitude, there shouldn't be a big difference in the average wage change. But as I showed when using OWE you vary easily get a negative change, so how could this be?
So I suggest that maybe if the most low-paying jobs are lost, then there could theoretically be an "increase" in wages without any real increase. So at least some of the "increase" measured in the OWE denominator isn't real, and when properly calculated (total wage change instead of average) the OWE would be higher.
The elasticity the entire time is OWE, though I can see how you could get confused, since I didn't explain the difference between MWE and OWE until near the end. So you don't need to make any adjustments to the calculations in section 5; they're already in terms of OWE
I understand, but according to my calculation using MWE I got -1.6% total wage change, so now I'm trying to understand why there's such a huge difference when using MWE vs OWE - this shouldn't be the case!
So I think the problem is that you assume that a 10% increase in avg wage means that you can calculate the increase as 98%*10%, when in reality if the increase in avg wage is due to low wage workers getting fired, then this is incorrect, because theoretically the increase can also be 98%*0%.
let me copy my other comment here so people understand what I'm talking about:
"a. -0.23 is not the effected workers elasticity, it usually includes majority of people earning more then the new MW. If only 33% of them are effected, then only 33% had an increase in wages.
b. The increase is not 10% for all effected workers, because many earn something between the old MW and the new MW, so for them the increase might only be 5% for example. You should use about 6-7% increase in effected wages for 10% MW increase.
a+b mean you should use 6-7% increase for the 33% effected workers, which would give you:
7%*(0.33-0.023)-100%*0.023 = -1.6%
an immediate decrease in wages even for 10% MW increase.""
The issue is that the elasticity of -0.23 already is an OWE. The MWE is a completely different number which I did not calculate. The OWE already accounts for the fact that not everybody is affected; if someone's wages don't rise as a result of the MW change, then they won't affect the wage change which is the denominator for MWE
From what I saw the number for MWE it should be vary close to the OWE. The difference isn't big enough to explain the problem.
I'm not sure about that, the calculation of the average change is usually not done on an individual basis, are you sure that this is how the calculation is done in practice?
Fascinating. Totally makes you wonder, what if the 'reduced proft' actually sparks innovation in efficiency instead of just cutting jobs?
My guess is the following:
1- the minimum wage the higher it is, the bigger the burden for businesses, employees and employ-seekers this last 2 via the employment costs for the first one
2- big businesses are more favoured by MW than smaller ones as long as the formers have a higher financial effort and could be more easily displaced or prevented from entry to the market, this logic also applies to skilled and/or experienced workers in comparison to those who are less although here is more variable because some arguably lesser titulations, based in academical-education system and titulation hierarchy consensus, have a higher chance of employment than supposedly higher ones
2.1- Usually youth have less experience than older people
3- MW positive correlations relating to higher employment doesn't really exist is only that thanks to other situations and policies they constraint it's bad effects basically we succed despite of not thanks to this