For a simple calculation that should suffice in a stressful case interview, multiply population by disposable income/capita (or GDP/capita if you're desperate) and multiply that by a percentage of income that the average person seems likely to spend on retail every year (e.g. 33%?).
i.e. 300m * $40k * 0.33 = 4 trn
- which actually gets you pretty close to the ~4.5 trn spent on retail in the US last year. You'd get 5 trn from GDP per capita figures (~$50k/p.c.), so that would be an alright guess as well.
I would bottom up calculation in the following way
Of cause this calculation would not give you an exact figure however probably you would have an order of magnitude
What do you think? Does it make sense?
1. define retail and unit of measurement, let's say overall spending on an annual basis.
2. MS= # people * disposable income/person/year * retail spending% + non-personal spending on retail + tourists
3. 3 segments of people: high income/ mid income/low income. Each has different disposable income and retail spending%.
4. assign percentages to non-personal retail spending and tourists.
Please let me know what you think.