DATA EXPLORATIONS
  • Data Science Lab
  • Lucid Analytics Project

Are People in More Expensive Countries Richer?

4/3/2017

 
Picture

This is part 2 of an analysis of costs of living and local purchasing power, in five hundred major cities around the world.

For part one, click here: Primary Drivers of Costs of Living, Worldwide
For part three, click here:  Are People in More Expensive Countries Richer?

The data was sourced from Numbeo.com, which hosts user-contributed data - current within the last 18 months.
The IPython Notebook for this project is available on github.

Is there a relationship between cost of living and local purchasing power, in cities around the world?

Note that Local Purchasing Power is a measure that *already* takes into account the cost of living. It is not a measure of absolute wages; rather, it describes the amount of "spending power" a person has, given both:
​
  • Their wage, and
  • The cost of goods and services in their area

If wages correlate perfectly to cost of living, then there should be *no* variance in local purchasing power.
If there is no variance in local purchasing power, then it should hold no relationship to cost of living.
(Wages should scale evenly with costs, thus disallowing cost of living from influencing local purchasing power).

We do, in fact, see a low degree of correlation in most regions in the world:
Regions with no or very weak correlation between cost of living and local purchasing power: 
  • NORTH AMERICA  * P-value: 0.925 
  • OCEANA * P-value: 0.812 
  • INDIA * P-value: 0.184 
  • MIDDLE EAST * P-value: 0.493 
  • AFRICA  * P-value: 0.778
  • LATIN AMERICA * P-value: 0.215​
​

And yet, for two regions in the world, there is a very high correlation:
  • Europe:
    • Correlation Strength: STRONG
    • P-Value: 0.0
    • R-Squared: 0.53
  • Asia:
    • Correlation Strength: MEDIUM
    • P-Value: 0.001 
    • R-Squared: 0.251

These relationships become immediately clear when we plot local purchasing power vs cost of living:
Picture
Picture
The blue line in each chart is our regression line, and the light blue bands are our 95% confidence intervals.

We see from these plots that:
  • There is a definite correlation between local purchasing power and cost of living
  • The more expensive a city is to live in, the *MORE* spending power the average person in that city enjoys.

In Europe:
  • For every 1% rise in the cost of living, purchasing power goes up by 1.15%.
  • A 1% rise increase in cost of living corresponds to a 2.17% rise in expected wages.
  • This relationship alone can explain over half of the variance (53%).

In Asia:
  • For every 1% rise in the cost of living, purchasing power goes up by 0.98%.
  • A 1% rise increase in cost of living corresponds to a 2.0% rise in expected wages.
  • This relationship alone can explain approximately one quarter of the variance (25%).

Why might this be so?

​North America, and Oceana contain only rich countries, which are economically similar. There is a relatively low degree of variance in both income, and cost of living between - or within - these countries. Africa and Latin America are comprised mostly of poorer countries with both low costs of living, and low incomes. While there is definite variance between the cost of living in, say, Chile, and Paraguay, there is still not a tremendous amount of variance between countries.

Generally speaking - and this is only a hypothesis - it may be true that wages and cost of living tend to scale proportionally for cities within a single country, while they do not necessarily scale proportionally between countries. Europe and Asia - more so than any other regions - are both home to economies with a huge amount of variance between countries.

Plotting the data in Tableau, this becomes quite clear:
​
  • Color indicates the total cost of living in a city (red is expensive, green is cheap).
  • Size indicates local purchasing power in a city (bigger is wealthier).

In Europe, we see a very strong correlation between size and color.
  • More expensive cities are wealthier, despite the higher costs of living.
In America, we see very little correlation between size and color.
  • Some cities are cheap, with wealthy people; some are expensive with poor people; some are rich and expensive, some are poor and cheap.


Next up - Part Three: "The World Through Whose Eyes?" - Costs of living around the world, relative to your country of residence.

Comments are closed.
Powered by Create your own unique website with customizable templates.
  • Data Science Lab
  • Lucid Analytics Project