The for-rent ads in the San Francisco Chronicle on Sunday, April 2, 1961, sound dreamy:

Luxurious spacious 4 rooms. Many walk-in closets. All Utilities included. $155 a month in the Marina.

Deluxe 3. Garbage disposal, wall-to-wall carpet. Garage included. $125 and up in Pacific Heights.

Spectacular Marine view. 1-bedroom, sun deck. $175 on Telegraph Hill.

Those were the good old days, right?

In fact, in the early 1960s, rents in San Francisco were rising by an average of about 6.6 percent each year. As it turns out, that’s the same annual rate they would later rise in the 1970s, and in the late 1990s, and in the mid-2000s. It’s the rate they’re rising today.

Track changes in the San Francisco rental market over more than half a century, and the pattern looks like this, according to Eric Fischer:

Fischer, a Bay Area resident who knows his way around microfilm, manually mined for-rent ads in the San Francisco Chronicle going back to the late 1940s to supplement more recent data housing costs in the city. Adjust the picture above for inflation, and real rents have gone up about 2.5 percent each year, give or take a few wobbles:

The stability of this trend is remarkable, given that a lot of other things changed in San Francisco over this time. The city’s population declined in the 1960s and ’70s. Rent control was introduced in 1979. There have been waves of urban renewal and successive tech booms.

The steady line, for one thing, points to the fact that San Francisco’s high housing costs have been building for a very long time (and can’t so neatly be pinned on all the latest tech wealth). A $2,500 studio in 2016 is the product of 60 years of housing costs outpacing inflation. Add up all of those 2.5 percent annual increases, and the cost of rent has quadrupled over this time.

Fischer’s chart also suggests the incredible difficulty of reversing this trend. In general, when the economy is doing well, rent rises. When it stumbles, rent falls. Similarly, when we add a lot of new housing, costs for it decline or slow down. When we don’t, those costs go up. And so, Fischer explains with the data he has amassed, we can pretty accurately predict how much rent will rise if we know three things: how many housing units exist in San Francisco, how many people have jobs, and how high their wages are.

“Housing, it appears, gets expensive either when people get paid more or there are more people getting paid,” he writes on his blog, “and more steeply than simple linear increase.”

That means if San Francisco wants the cost of housing to go down, there are three clear ways to get there. The city could build more of the stuff. Or it could hold out for falling incomes or job losses among the people who compete for housing. The second two plans seem counterproductive (although there may in fact be frustrated residents rooting for those strategies).

If San Francisco wanted to revert to what rents were in the early 1980s, according to Fischer’s model, the city would need to add another 200,000 housing units. Or the people currently living there could take massive pay cuts — the equivalent of a 44 percent drop in their salaries. Or it might do the trick if half of all workers in San Francisco lost their jobs.

The alternatives make building more seem even more imperative. Or, to put that better:

San Francisco can have a dynamic economy and charming neighborhoods unmarred by new construction and denser housing. But it can’t have both of those things without paying a steep cost in rent (and without pushing lower-wage workers out). Other cities face a similar fate if their economies boom but their housing construction does not.