Trimming your dataset to start at 2012 hides the fact that that was the trough of a massive crash that brought housing prices well below wages, and dramatises the growth in prices when for years they were basically just catching up with wages.
The original Oireacteas report which used the original graph mentions this as a caveat. The report’s use is not misleading – it is trying to show the 10 year trend but makes it clear that the prices of 2012 were anomalously low. Using this graph without explaining this context is misleading.
The trend of housing prices and rent (especially since 2020) are clearly bad. But you don’t need to use misleading graphs to make that point.
One interesting thing the extended graph really shows is that whatever about the property crisis, the *rental* crisis is particularly horrendous. The gulf between rising rents and wages has never been this bad, not even at the peak of the housing bubble.
I didn’t go to the effort of trawling through the datasets used in the original Oireachteas report because I didn’t want to devote a day to making this and the St. Louis Fed site has great interfaces for trimming and normalising its datasets.
The biggest discrepancy is the wages – the Oireachtas dataset is broader, in some way accounting for weekly wages and low hours, as well as including public sector workers. An admittedly tiny bit of research makes me think public and private sector wages will come out as a wash though considering private sector has both some really high salaries and some horrendously low hourly wages.
Accounting for hours really makes the wage datasets of the two charts very difficult to directly compare, especially because the Oireachtas report doesn’t describe how the multiple datasets are amalgamated to make the line they call “Wages”. ¯\_(ツ)_/¯
**Appendix 2: I’m an idiot**
I mixed up the colours used in the first graph when making my graph. My graph’s dataset labels are still correct. Sorry if this confused anyone.
Man I should of had the foresight to buy a house in 2012-2014, too bad I was busy being a child,
Hmmmmm I love the smell of contextual thinking in the morning.
So we’re like in 2001 but with high-speed internet everywhere?
5 comments
I don’t mean to downplay the severity of the housing crisis, but [the use of this graph in isolation](https://www.reddit.com/r/ireland/comments/1715ktr/growth_in_wages_rent_and_property_prices_between/) is a textbook example of how to mislead with statistics.
Trimming your dataset to start at 2012 hides the fact that that was the trough of a massive crash that brought housing prices well below wages, and dramatises the growth in prices when for years they were basically just catching up with wages.
The original Oireacteas report which used the original graph mentions this as a caveat. The report’s use is not misleading – it is trying to show the 10 year trend but makes it clear that the prices of 2012 were anomalously low. Using this graph without explaining this context is misleading.
The trend of housing prices and rent (especially since 2020) are clearly bad. But you don’t need to use misleading graphs to make that point.
One interesting thing the extended graph really shows is that whatever about the property crisis, the *rental* crisis is particularly horrendous. The gulf between rising rents and wages has never been this bad, not even at the peak of the housing bubble.
**Sources:**
[1]: [Private Sector Labour Compensation (hourly)](https://fred.stlouisfed.org/series/LCEAPR01IEQ661N)
[2]: [Residential Property Prices](https://fred.stlouisfed.org/series/QIEN628BIS)
[3]: [Housing Affordability for Private Household Buyers in Ireland (original graph source)] (https://data.oireachtas.ie/ie/oireachtas/parliamentaryBudgetOffice/2023/2023-10-05_housing-affordability-for-private-household-buyers-in-ireland_en.pdf)
**Appendix 1: Different Datasets**
I didn’t go to the effort of trawling through the datasets used in the original Oireachteas report because I didn’t want to devote a day to making this and the St. Louis Fed site has great interfaces for trimming and normalising its datasets.
The biggest discrepancy is the wages – the Oireachtas dataset is broader, in some way accounting for weekly wages and low hours, as well as including public sector workers. An admittedly tiny bit of research makes me think public and private sector wages will come out as a wash though considering private sector has both some really high salaries and some horrendously low hourly wages.
Accounting for hours really makes the wage datasets of the two charts very difficult to directly compare, especially because the Oireachtas report doesn’t describe how the multiple datasets are amalgamated to make the line they call “Wages”. ¯\_(ツ)_/¯
**Appendix 2: I’m an idiot**
I mixed up the colours used in the first graph when making my graph. My graph’s dataset labels are still correct. Sorry if this confused anyone.
Man I should of had the foresight to buy a house in 2012-2014, too bad I was busy being a child,
Hmmmmm I love the smell of contextual thinking in the morning.
So we’re like in 2001 but with high-speed internet everywhere?
Why are the wages going up and down all the time?