Security & Order: Shared Priorities, Different Emphasis

Military: the bipartisan baseline is long gone

Of the three security-and-order items the GSS tracks, military spending has the most volatile history. The proportion of Americans saying the government spends "too little" on the military starts at 12% in 1973 — the immediate post-Vietnam moment — spikes to 60% in 1980 (the Reagan election year, with the Soviet invasion of Afghanistan and the Iran hostage crisis both fresh), collapses to 11% in the 1993 post-Cold-War trough, and has climbed gradually since to 30% in 2024.

What the topline hides is that the 30% figure for 2024 is held up almost entirely by one party.

Two things stand out in Figure 2.

First, the parties used to agree on the military. Through 1990, Democratic and Republican readings on military spending were generally within a few points of each other, and on either side of the Reagan-era peak the partisan order was sometimes reversed — slightly more Democrats than Republicans wanted bigger defense budgets in 1973 (12% vs 10%). The contemporary identification of military spending as a Republican-coalition issue is a post-2000 phenomenon, not an inherent feature of American politics.

Second, the divergence is one-sided since the 1980 spike. Outside that single Reagan-election-year wave (when both parties surged together, Democrats to 59% and Republicans peaking similarly), the Democratic line on military spending has held in the 9–29% range. The Republican line is what changed: from 10% in 1973 to 50% in 2024, with the sharpest rises in the 2000s (Iraq War era) and the post-2014 period. The 36-point gap exists because Republicans moved about 40 points across the long arc and Democrats moved roughly two.

The age structure of military-spending support has also changed in a way the topline conceals.

The 1973 reading shows an absence of generational divide — every age group is in the low teens, all wanting roughly nothing in additional defense spending. The 2024 reading shows a clean age gradient, with each successively older cohort more supportive than the last. The 65+ cohort sits at 46%; the 18–34 cohort at 19%. The cohorts whose adulthood was structured by the Cold War, the post-9/11 wars, and the bipartisan-defense-budget norm of the 1980s have remained pro-military. The cohorts whose adulthood spans Iraq, Afghanistan, and the post-2014 fragmentation of Republican views about NATO and the alliance system are notably more skeptical.

When you cross the age structure with the partisan structure, the contemporary military hawk is not a generic Republican — it is an older Republican. Younger Republicans are still well above younger Democrats, but the largest concentrations of "spend more on the military" sentiment sit specifically in the 65+ Republican cell.

Crime: the parties have switched places

If military spending shows a partisan story where one coalition moved and the other held still, crime spending shows something stranger: both coalitions moved, but in opposite directions, and they passed each other in the middle.

The crime-spending topline is essentially flat over fifty years. About two-thirds of Americans, in nearly every wave, want the federal government to spend more on reducing crime. It is one of the most bipartisan items in the GSS spending battery — and looking only at the topline, you would describe it as the kind of issue both parties agree on. The party-broken-out version says something more interesting.

For most of the GSS's history, Democrats were modestly more supportive of federal crime spending than Republicans. This was the era of the 1994 crime bill, the Democratic embrace of community policing, and the broad consensus — across both parties — that the post-1960s crime wave required federal action. The 1990 reading shows Democrats at 73% and Republicans at 67%; the lead was small but persistent.

The contemporary picture is the inverse. By 2024 Republicans stand at 76% and Democrats at 61% — Republicans are now the more spend-on-crime-reduction coalition, with a 15-point gap. The lines cross sometime around 2018, in the period during which "defund the police" became a salient Democratic-coalition phrase and through which the post-2020 rise in urban violent crime became a salient Republican-coalition concern. Whatever happened in this period, it did not just widen a pre-existing gap. It flipped the direction of one of the most stable spending preferences in American politics.

The educational-divide story on crime is comparatively muted. The college / non-college gap was small in 1973 (3 points), widened in the 2000s (when college-educated Americans were notably less likely to say crime spending was too low — a Bush-era civil-liberties dynamic), and has compressed back to 5 points in 2024. Whatever the partisan-realignment story is, it is not primarily a story of educational sorting. It is a story of two coalitions whose ranks happened to contain different people in different decades, and whose median crime-policy preferences inverted as the coalitions reshuffled.

Drug addiction programs: fifty years of nearly unchanged consensus

The last item in this section is the most boring in the GSS's spending battery — and in a chapter on polarization, that is its interest.

The drug-spending topline is essentially flat across the entire history of the GSS, with two small bumps and one small dip: 70% in 1973, peak of 73% during the late-1980s crack era, trough of 56% in the early-2000s prescription-pill years before the opioid crisis went mainstream, and back to 68% in 2024. The opioid epidemic — the defining American drug story of the past two decades — barely registers in this series. The country's attitude toward federal drug-program spending has been "we should be spending more" for fifty years, in every decade, with only modest variation.

The party-broken-out version is similarly stable.

A 9-point partisan gap on a 68% topline is a domain where the parties effectively agree. Of the three security-and-order items in this section, drug-addiction programs are the one a contemporary political strategist would look at and conclude: both coalitions want the same thing here, even when their voters disagree about everything else. The Democratic line is consistently a bit higher; the Republican line tracks it; both coalitions, in every decade, contain a clear majority that wants more federal spending on the problem. The post-2014 bipartisan criminal-justice-reform moment, the opioid settlements with Purdue Pharma, the surprisingly cross-party legislative work on treatment access — these all show up downstream of what the GSS has been measuring for fifty years.

"Shared priorities, different emphasis" — half right

The proposal that became this book described §3.3 as "Security and Order: Shared Priorities, Different Emphasis." On the evidence in this chapter, that framing is half right.

The "shared priorities" part is correct on crime and drugs. Both coalitions, across both contemporary readings, want federal spending on crime reduction (61% Democrats, 76% Republicans in 2024) and on drug-addiction programs (72%, 63%). A political analyst looking only at the topline support figures would correctly conclude that these are issues with broad cross-party majorities. Whatever the parties disagree about on how the federal government should spend on safety, they agree, in their bases, that more federal spending on safety is desirable.

The "shared priorities" part is wrong on the military. The 36-point partisan gap on defense spending is on the same order of magnitude as the largest partisan gaps in the entire GSS spending battery — comparable to the gap on programs explicitly named "improving the conditions of Black Americans" (§3.2, 46 points), and larger than the gap on welfare (34 points). The military is now a partisan-coded category, in a way it was not for the first thirty years of the GSS. Treating it as a shared priority belongs to the era of the Reagan defense buildup and the post-9/11 consensus. Both of those eras are over.

And the "different emphasis" part, on crime, misses what actually happened. The two parties did not simply disagree on how to reduce crime; they switched which party was the more enthusiastic spender on it. Democrats were the law-and-order coalition in the 1970s; Republicans are in the 2020s. The framing of "Democrats want to attack root causes, Republicans want to fund police" is contemporary and roughly accurate for present-day policy disputes — but it is not the same coalition that supported the 1994 crime bill. The voters who responded to that bill's pro-police orientation and to the Bush 41 administration's tough-on-crime rhetoric were disproportionately Democrats then; their grandchildren and their political successors are disproportionately Republicans now.

The next chapter turns to a smaller but no less revealing set of spending domains — the environment and the country's physical infrastructure — where a different kind of sorting is now underway.