3 ms·
> When property values increase, so do rents. Maybe, but probably not at the same rate. Broadly speaking, rents are tied to incomes, house prices to the avail
by jbarham 8y ago
> When property values increase, so do rents.
Maybe, but probably not at the same rate.
Broadly speaking, rents are tied to incomes, house prices to the availability and cost of credit (i.e., interest rates). Average income places an upper bound on rents because renters can only pay rent from income (since no bank will lend you money to pay rent).
I'm currently benefiting from this phenomenon myself as I rent a house for ~$25k/year (in Melbourne, Australia). Buying that same house would cost me ~$45k/year.
- thaumasiotes 8y ago> no bank will lend you money to pay rent I'm pretty sure every bank will do this, in the form of a credit card.
- alsetmusic 8y ago>> no bank will lend you money to pay rent >I'm pretty sure every bank will do this, in the form of a credit card. I’ve never seen an option to pay rent with a credit card. Cash, check, money order, or cashier check have been options I’ve seen.
- thaumasiotes 8y agoCredit cards will advance you cash that you can use for whatever you want.
- namibj 8y agoWhich bank allows you to take >10 years to pay it off? For less than 5% (over inflation) annual interest?
- dcosson 8y ago> Broadly speaking, rents are tied to incomes, house prices to the availability and cost of credit Mind citing a source for this? I've seen a situation where rents and house prices didn't both rise together. Maybe not precisely the same rate, there may be some noise from other factors, but within a few percentage points of each other. Rents may be somewhat related to the incomes of the people who are renting in the area, but these people are not necessarily the average population. In a housing shortage the people with high paying jobs rent decent places, everyone who makes less is forced to live with roommates or commute from further away and many leave the area altogether. And then in your example, that's not necessarily so crazy. Some of that $45k goes towards the mortgage interest which is tax deductible (in the US, not sure about Australia), and of course over time more and more of it goes to the principle so effectively paying yourself.
- jbarham 8y agohttps://www.economist.com/graphic-detail/2018/02/09/the-economist-house-price-indices https://www.economist.com/graphic-detail/2018/02/09/the-econ... lets you play around with comparing house prices vs income and rents. In Australia (where I live) and Canada (where I'm from) house prices are more overvalued compared to rent than income (i.e., rents have increased slower than house prices). > Rents may be somewhat related to the incomes of the people who are renting in the area, but these people are not necessarily the average population. By definition renters must live in the are where they rent, and must work within commuting distance of where they rent (even if it's a very long commute). Landlords, however, can live anywhere, provided they have someone to manage their rented properties. E.g., my landlord lives in Singapore, but I pay my rent to a Melbourne based property management agency. But since I live (and work) in Melbourne my rent is set by the market rate in Melbourne, not the (much higher) rate in Singapore. Unfortunately for non-techies in SF, their rents have been pushed up massively due to the much higher tech salaries. But that only proves the strong correlation between rent and income. BTW, mortgage interest in not tax deductible in Australia. Somewhat crazily, however, rental losses on investment properties are tax deductible ("negative gearing").