Presidential approval: It’s the economy; except when it’s not

The “Carville Dictum” at last is getting a fresh look — and that’s a good thing for all those interested in the interplay between presidential approval and the economy.

The longstanding misinterpretation is not political strategist James Carville’s fault. He correctly identified the weak economy in 1992 as the root of then-President George H. W. Bush’s unpopularity and Bill Clinton’s path to the presidency.

{mosads}But this lesson has morphed over time into a vastly oversimplified formula: Bad economies make presidents unpopular; good economies make them popular.

Donald Trump proves otherwise. But he’s not the first. 

Even if not unprecedented, the president’s position is striking. Consumer sentiment, as measured in the weekly Bloomberg Consumer Comfort Index, is at a 17-year high; unemployment is at a 17-year low; and the bull market plows ever forward.

Yet Trump’s job approval rating is the lowest on record for any president heading into his first midterm elections at least since 1954. 

What’s most notable about the president’s poor score in a strong economy is that he doesn’t have an unpopular war to blame it on.

Harry Truman suffered low approval at a time of high employment from 1950 through 1952 as U.S. troops slogged through the Korean War. So did Lyndon B. Johnson in 1967-68 as the Vietnam War deepened and George W. Bush in 2007 as the war in Iraq did the same.

Beyond demonstrating that it needn’t take a war gone ugly for a president to fumble, Trump breaks the mold in another way: He was deprived of the honeymoon presidents typically enjoy at the start of their terms in office, regardless of economic conditions. This president began low and has stayed there.

Yet molds are made to be broken. We’ve tested the relationship between presidential approval and the economy back to 1948, using the unemployment rate as our indicator of economic conditions, given its direct impact on the public.

We find that employment does interact with presidential approval. But when and how much is highly variable. Some presidents’ job approval ratings have correlated strongly with employment levels. Others’ have had little to no relationship with the employment rate or even have moved in the opposite direction than expected.

This shouldn’t be a surprise. Americans use a range of factors in assessing presidents — the economy for sure, but also performance on policy, response to national and world events and personal characteristics.

Even without the influence of war or economic conditions, plenty of past presidents have suffered from scandals and setbacks or gained from successes.

We conducted our analysis by splitting unemployment into quartiles:

• Tier 1 represents periods of unemployment of 4.5 percent or lower, generally considered full employment;
• Tier 2, 4.6 to 5.5 percent;
• Tier 3, 5.6 to 6.8 percent; and
• Tier 4, 6.9 percent unemployment or higher. 

We’re easily in a first-tier economy, yet Trump has just 36-percent approval in the latest ABC News/Washington Post poll.

As noted though, he’s had company: George W. Bush had a first-tier economy in early 2007, but it was coupled with approval as low as 35 percent. LBJ dropped as low as 35 percent approval with a Tier-1 economy in 1968. And Truman’s approval rating hit a dismal 22 percent despite Tier-1 conditions.

Correlations describe the relationships among variables: 1 when they move precisely together, -1 when they move precisely opposite, 0 when their movements are entirely unrelated.

Looking at presidents across their terms, we see high correlations between approval and unemployment for Gerald Ford (.70), George H.W. Bush (.71), Dwight Eisenhower (.68) and Clinton (.66).

Yet approval of others has had little to no relationship with employment or has even moved in the opposite direction than expected: Truman (-.65), Johnson (-.90), John F. Kennedy (-.49) and now Trump (-.66).

Kennedy, Jimmy Carter, Ronald Reagan and Barack Obama all started with high approval — and for a time held it — despite Tier-4 unemployment; their predecessors, at least for a time, took the rap.

It’s easy to spot the influence of non-economic factors. Nixon plummeted to an approval rating in the 20s, courtesy of Watergate, despite a Tier-2 economy. George W. Bush soared in approval after 9/11 and stayed high for nearly two years; his response to terrorism, not the Tier-3 economy, commanded public attention. 

The conclusions are straightforward: A bad economy makes it especially difficult for a president to be popular. A better economy doesn’t produce popularity — it just makes it more achievable. And the Carville Dictum needs a simple revision: It’s the economy, stupid. Except when it ain’t.

Gary Langer is the president of Langer Research Associates. Research Associate Allison de Jong contributed to the writing of this piece.