The “third rail” of American politics is certainly well-known. Social Security — along with Medicare and Medicaid, the other major entitlement programs — has served as the “essential” government program for more than 70 years. It serves as a lifeline for Americans older than 65, raising an estimated 14 million of them above the poverty line. And — because it’s essential to so many, but also so woefully expensive and inefficient — it’s in desperate need of reform.

From the very start, Social Security was intended to be a more limited program than the one we have today. Back in 1937, the average life expectancy was just 58 years — the entitlement program as first conceived was catering to a small subset of the population who, even if they had reached 65, likely wouldn’t live much past it. Today, the average American can expect to live almost 79 years, and those who reach age 65 can expect to live another 19 years for men and another 21 for women. While there were 4.9 workers per Social Security beneficiary in 1960, that ratio was 2.8:1 as of 2010 and is projected to further deteriorate to 1.9:1 in 2035.

The impact of those demographic trends on the viability of Social Security going forward is both substantial and concerning. Spending on Social Security and Medicare currently totals around $1.34 trillion, or 38 percent of federal expenditures. The Social Security program has operated at a loss since 2010, and will continue to do so — producing ever larger deficits — into the foreseeable future. By 2038, the program will have exhausted its trust funds altogether.

Commonsense reforms are a good place to start. Raising the retirement age from 65 to 70 in a staggered manner until 2040, for example, would reduce total Social Security outlays by 6 percent over current projections by that year. Reducing cost-of-living adjustments by 0.5 percentage points per annum would reduce total outlays by 7 percent by 2040, and employing the chained consumer price index measurement of inflation could attain a similar effect.

Being smarter about who receives benefits is equally valuable. Social Security was originally intended as a means of lifting the elderly out of poverty during the Depression era. That objective still has exceptional merit for those 14 million Americans who rely on Social Security to lift them out of poverty, but means-testing the program — reducing benefits to those who, quite frankly, don’t need them — could extend the life of the trust for years, according to the Congressional Budget Office.

Of course, such reforms are at best marginal in impact, postponing rather than averting altogether the fiscal reckoning. Hiking taxes to pay for projected entitlement spending increases — including those expected for Medicare and Medicaid — would likely impose a burden on affluent Americans beyond that already exacted by the OECD’s most progressive tax system. Instead, it makes more sense to fundamentally reassess what Americans receive from a fundamentally unsustainable program.

That reassessment should start with transitioning from seeking to replicate income levels — unsustainable given rising personal incomes and additionally damaging by discouraging savings — to providing a basic supplement that averts poverty throughout retirement, and with supporting initiatives that promote independent fiscal responsibility and savings by workers today by incentivizing them to keep tomorrow in mind too.

Social Security, by most accounts, will not be accountable for most of the projected growth in entitlements spending. From the Affordable Care Act to Medicare and Medicaid, America has become accustomed to programs that simply aren’t sustainable over the long run. Touching the third rail of American politics — and living to tell the tale — is, however, a good place to start.

Contact Marshall Watkins at mtwatkins@stanford.edu.