9.35 Myths about Durability of the Compact

Tara Ross, a lobbyist against the National Popular Vote Compact who works closely with Save Our States, asserted in 2017:

“States will be able to pop in and out of the compact. … The presidential election system would be in a constant state of upheaval.”[970]

In a similar vein, David Gringer propounded a hypothetical scenario in 2008 in which the Republican-controlled Texas legislature might “perniciously” enact the National Popular Vote Compact on the basis of mid-year polling indicating that the Republican presidential nominee is poised to win the national popular vote while losing the Electoral College.

In his 2008 article, Gringer hypothesized a 2020 election cycle in which:

“Early polling shows likely Democratic nominee New York Governor Eliot Spitzer with a substantial lead in Texas over the soon-to-be Republican nominee South Dakota Senator John Thune. If the Democratic nominee carries Texas in the general election, he will have a ‘lock’ on the electoral college.”

“At the behest of Republican Party leaders, the state legislature passes a bill awarding its electoral votes to the winner of the national popular vote. The Republican Governor of Texas signs the bill into law.

“With the addition of Texas, enough states now participate for the NPV to take effect.”[971]

The Compact governs a given presidential election only if it is in effect in states possessing a majority of the electoral votes on July 20 of a presidential election year.

The National Popular Vote Compact specifies that it will:

“govern the appointment of presidential electors in each member state in any year in which this agreement is, on July 20, in effect in states cumulatively possessing a majority of the electoral votes.”[972]

Moreover, if a state withdraws from the Compact during the six-month period between July 20 of a presidential election year and the following January 20 (Inauguration Day), the withdrawal will not take effect until after the Inauguration (as discussed in detail in section 9.25).

Thus, the question of whether the National Popular Vote Compact will govern a particular November presidential election is settled on July 20 of each presidential election year.

July 20 usually comes before events such as:

• the national nominating conventions of the major political parties;
• the debates between the candidates nominated at the conventions; and
• the day-to-day conduct of the campaign in the fall, including the numerous unexpected events that occur (e.g., the financial crisis in mid-September 2008).

Polls taken days before an election have consistently failed to accurately foresee that the electoral vote will diverge from the national popular vote.

Hypothetical scenarios in which politically motivated states “pop in and out of” the Compact before July 20 of a presidential election year assume the existence of predictive techniques that are sufficiently accurate to convince a state legislature and Governor that the Electoral College vote will diverge from the national popular vote in November.

Both the 2000 and 2016 election illustrate that it is difficult to make an accurate prediction days before Election Day—let alone in July—about whether the electoral vote will diverge from the national popular vote.

In 2000, for example, George W. Bush had an average national-popular-vote lead of 2.1% in the 18 polls listed in Polling Report for the five-day period before Election Day. As shown in table 9.52, Bush was ahead in 14 of the 18 polls and tied in two, while Gore was ahead in only two polls.[973]

Table 9.52 National polls in the five days before the 2000 election

Poll	Bush	Gore	Nader	Buchanan	Bush lead over Gore	Day
CBS *	44	45	4	1	-1	Monday
CNN/USA TODAY/GALLUP # @	48	46	4	1	2	Monday
CNN/USA TODAY/GALLUP #	47	45	4	1	2	Monday
IBD/CSM/TIPP * # @	47.9	46	3.7		2	Monday
REUTERS/MSNBC * # @	46	48	5	1	-2	Monday
VOTER.COM * # @	50	45	3.5		5	Monday
VOTER.COM * #	46	41	4	0	5	Monday
ABC # *	48	45	3	1	3	Sunday
HARRIS * @	47	47	5		0	Sunday
HOTLINE *	45	42	4	1	3	Sunday
ICR *	46	44	7	2	2	Sunday
NBC/WALL ST. JOURNAL	47	44	3	2	3	Sunday
PEW @	49	47	4		2	Sunday
PEW	45	43	4		2	Sunday
WASHINGTON POST *	48	45	3	1	3	Sunday
FOX/OPINION DYNAMICS *	43	43	3	1	0	Thursday
MARIST COLLEGE	49	44	2	1	5	Thursday
NEWSWEEK	45	43	5		2	Thursday
Average	46.7	44.6	4.0	0.7	2.1

Mike Shannon reported what the Bush campaign thought six days before Election Day:

“I was the keeper of the George W. Bush campaign map—our color-coded projection of electoral college votes based on our private state tracking polls.”

“The map is a prime example of the uncertainty in projecting presidential raceswhen there is a deeply divided electorate and the potential for late-fall surprises.”

“That map proved wrong just six days later.”

“Florida, where our nightly polls had shown us holding a five-point lead.”

“New Mexico (Solid Bush), Michigan (Lean Bush) and Wisconsin (Lean Bush) were ultimately won by Al Gore.”

“My conviction … can be summed up in three words: Nobody knows anything.”[974] [Emphasis added]

Two days before Election Day (Sunday November 5, 2000), the lead story on the front page of the New York Times said:

“Mr. Bush continued to hold leads ranging from one to five percentage points in all the national tracking polls, in addition to his slight advantage in the electoral college calculations.”

“No fewer than a dozen states, with a total of 125 of the 270 electoral votes needed for election, were classified as tossups by politicians, pollsters and academic specialists interviewed by the New York Times.”[975] [Emphasis added]

Meanwhile, in 2000, the Bush campaign was actively planning for the possibility of losing the Electoral College while winning the national popular vote.

In an article entitled “Bush Set to Fight an Electoral College Loss,” the New York Daily News reported on Wednesday November 1, 2000:

“Quietly, some of George W. Bush’s advisers are preparing for the ultimate ‘what if’ scenario: What happens if Bush wins the popular vote for President, but loses the White House because Al Gore won the majority of electoral votes?”

“‘The one thing we don’t do is roll over,’ says a Bush aide. ‘We fight.’

“How? The core of the emerging Bush strategy assumes a popular uprising, stoked by the Bushies themselves, of course.

“In league with the campaign—which is preparing talking points about the Electoral College’s essential unfairness—a massive talk-radio operation would be encouraged. ‘We’d have ads, too,’ says a Bush aide, ‘and I think you can count on the media to fuel the thing big-time. Even papers that supported Gore might turn against him because the will of the people will have been thwarted.’

“Local business leaders will be urged to lobby their customers; the clergy will be asked to speak up for the popular will; and Team Bush will enlist as many Democrats as possible to scream as loud as they can. ‘You think ‘Democrats for Democracy’ would be a catchy term for them?’ asks a Bush adviser.

“The universe of people who would be targeted by this insurrection is small—the 538 currently anonymous folks called electors, people chosen by the campaigns and their state party organizations as a reward for their service over the years.”

“Enough of the electors could theoretically switch to Bush if they wanted to—if there was sufficient pressure on them to ratify the popular verdict.”[976] [Emphasis added]

Nate Cohn wrote in 2012:

“There is a high evidentiary burden for demonstrating that any candidate holds a structural advantage in the Electoral College. The Electoral College almost always follows the popular vote, and even when the popular vote winner fails to secure the necessary electoral votes, it isn’t necessarily apparent in advance. Heading into Election Night 2000, the fear was Gore winning the Electoral College and Bush winning the popular vote. The exact opposite happened only a few hours later. In an extremely close national election, deviations of only a few percentage points in the closest few states can complicate even the best gamed electoral scenarios.”[977] [Emphasis added]

Similarly, when the polls closed on Election Day in 2016, it was not evident that Donald Trump was going to win the Electoral College while losing the national popular vote. That outcome did not become evident until several hours later in the evening.

Thus, in the two elections of the early 2000s when the winner of the electoral vote diverged from the national popular vote, that outcome was not evident days—or even hours—in advance.

Summer polling cannot accurately predict that the electoral vote will diverge from the national popular vote in November.

In an article entitled “Anything Can Change in a Presidential Year,” Larry Sabato cited numerous previous inaccurate summer predictions:

“In June 2004, Kerry led Bush outside the margin of error at 49 percent to 43 percent. Instead, Bush grabbed his second term with 51 percent in November.”

“John McCain actually led Barack Obama by a whisker in Gallup’s daily tracking at the beginning of June 2008, 46 percent to 45 percent. It wasn't close in the fall, with Obama winning 53 percent.”[978]

In July 1988, Michael Dukakis led George H.W. Bush by 17% in a national Gallup poll and even led Bush by 10% in Bush’s home state of Texas.[979] However, the general-election campaign had not even started in July, and Bush won in November with an 8% national lead.

In June 2016, an ABC News-Washington Post poll of registered voters nationwide showed Hillary Clinton with a 12-point lead over Trump.[980]

In June 2020, a New York Times-Siena College poll showed Biden leading Trump by 14 points.[981]

Similarly, Timothy Stanley reminded us that:

“In March 1980, President Jimmy Carter led Ronald Reagan by 25 percent in some polls. Reagan went onto win the November election by 51 to 41 percent.”[982]

The New York Times reported that a nationwide Gallup poll taken on June 4–8, 1992, showed Bill Clinton in third place. The results were:

• Ross Perot—39%
• Incumbent President George H.W. Bush—31%
• Bill Clinton—25% support.[983]

Despite this June 1992 poll, Bill Clinton took the lead immediately after the Democratic convention and retained it all the way to Election Day.

Table 9.53 shows a compilation by Nathaniel Rakich of the FiveThirtyEight national polling average 84 days before 12 recent elections (that is, mid-August).[984]

Table 9.53 Mid-August national polling averages 1976–2020

Year	August leading candidate	Lead of August leading candidate in the national popular vote	Did August leading candidate still lead in the national popular vote on Election Day?	Actual lead of August leading candidate on Election Day	Absolute value of difference
2020	Biden	+8.3%	Yes	+4.0%	4.3%
2016	H. Clinton	+6.6%	Yes	+2.0%	4.6%
2012	Obama	+0.5%	Yes	+3.9%	3.4%
2008	Obama	+2.6%	Yes	+7.2%	4.6%
2004	Kerry	+2.5%	No	–2.4%	.4,9%
2000	G.W. Bush	+10.0%	No	–0.5%	10.5%
1996	B. Clinton	+11.3%	Yes	+8.5%	2.8%
1992	B. Clinton	+20.1%	Yes	+5.6%	14.5%
1988	Dukakis	+5.6%	No	–7.8%	13.4%
1984	Reagan	+16.0%	Yes	+18.2%	-2.2%
1980	Reagan	+22.1%	Yes	+9.7%	12.4%
1976	Carter	+26.6%	Yes	+2.1%	24.5%
	Average				5.5%

More importantly, presidential elections in which one candidate wins the popular vote while losing the electoral vote are necessarily close elections.

Tilden’s 3.0% margin in 1876 was the largest difference in the national popular vote among the nation’s “wrong winner” elections (table 1.1).

In 2000, the difference in the national popular vote between the two candidates was 0.5% (about a half million votes nationwide).

An article on July 24, 2012, by Nate Silver in the New York Times, entitled “State and National Polls Tell Different Tales About State of Campaign,”[985] reinforces the point.

Silver pointed out that the Real Clear Politics average of national polls gave President Obama a nationwide lead of 1.3% on July 24, 2012. However, at the same moment, Obama led by a mean of 3.5% in the Real Clear Politics averages for 10 battleground states (Ohio, Virginia, Florida, Pennsylvania, Colorado, Iowa, Nevada, Michigan, New Hampshire, and Wisconsin) that were considered (at the time) to be most likely to determine the outcome of the 2012 election. Both the 1.3% margin and the 3.5% margin cited in Silver’s article were inside the margin of error for most political polling taken during campaigns.

In any case, a poll is not a prediction. Even if a given political poll had no margin of error (e.g., if every voter in the country were polled), it would merely be a snapshot of public opinion as of the particular moment when taken.

State constitutional provisions and state legislative procedures provide numerous tools by which the minority party in a state legislature can frustrate a politically motivated last-minute change in state law.

As a practical matter of state legislative scheduling, a decision to “pop in or out of” the National Popular Vote Compact would have to be made considerably earlier than July 20 of a presidential election year in most states.

For one thing, a large majority of state legislatures are not even in session in mid-July. Thus, a change in a state’s method of awarding electoral votes would, in practice, have to be enacted earlier in the year in most states.

Moreover, some state legislatures (including the Texas legislature specifically discussed by Gringer) only meet in regular session at the beginning of each odd-numbered year—that is, about a year and a half before the presidential election.

In every state, winning approval of any new state law is a multi-step process that can be derailed at many points.

Moreover, the date of approval of a new state law should not be confused with the date on which the new law takes effect.

The National Popular Vote Compact is based on state laws that are in effect on July 20. Specifically, the ninth clause of Article III of the Compact provides:

“This article shall govern the appointment of presidential electors in each member state in any year in which this agreement is, on July 20, in effect in states cumulatively possessing a majority of the electoral votes.” [Emphasis added]

A new state law that is passed by the legislature and signed by the Governor can be “in effect” by July 20 only in accordance with the state’s constitution schedule specifying when new state laws take effect.

Procedures exist in every state legislature to allow a newly enacted law to take effect immediately. However, in many states, these procedures can be invoked only by a super-majority.

Given that the premise of Gringer’s hypothetical scenario is that the majority party in Texas (the Republicans) wants to enact the National Popular Vote Compact for a “pernicious” partisan advantage, the Democrats in the legislature would vigorously employ every available dilatory tactic at their disposal to block the bill.

Texas is one of four states with a two-thirds quorum in the legislature. Texas Republicans have never had a two-thirds super-majority in both houses of the legislature.

In 2003, the Democrats pulled the quorum when the Republicans attempted to pass a politically motivated mid-decade redrawing of the state’s congressional districts (section 9.25.1). They did so again in 2021 concerning an abortion bill.[986] They would surely do so in the situation envisioned by Grainger.

Moreover, many state constitutions provide for a significant delay between the time a Governor signs a new law and its effective date. Delays of 60, 90, or 120 days are typical (as shown in table 9.40 and discussed in section 9.25.1).

Looking at Texas in particular, new state laws take effect 90 days after enactment. This 90-day delay can only be waived by a two-thirds vote of both houses of the legislature. Texas Republicans have not had a two-thirds super-majority in both houses of the state legislature at any time in the 21^st century. Thus, the National Popular Vote Compact would have to be signed into law, at the minimum, by April 20 of a presidential election year in order to be in effect by July 20.

There is an additional reason why Gringer’s hypothetical scenario could not be successfully executed by April 20 of the presidential election year.

The Texas legislature is one of a number of state legislatures that meet only for a few months in odd-numbered years. Thus, Gringer’s hypothetical scenario would have to be executed in Texas during the short biannual session that takes place in the early part of an odd-numbered year—that is, 18 months before a presidential election.

Of course, it is theoretically possible to pass a bill in a special session of a state legislature. However, in Texas, a special session would not be an option for a highly partisan bill. If a special session were called for the purpose of passing an elections bill that is perceived to be of partisan advantage to the Republicans, Texas Democrats will simply prevent the formation of a quorum for the special session. In states with filibusters, the minority opposing the legislation would employ that tactic.

A state legislature’s dominant party can usually overcome dilatory tactics such as quorum-pulling in the longer regular legislative session when there are numerous “must pass” bills. However, in a short special session when only one bill is on the agenda, quorum-pulling and filibustering are generally highly effective.

Section 9.25.1 provides additional details on the difficulties associated with trying to pass legislation over the determined opposition of a legislature’s minority.

In short, Tara Ross’ assertion that: “States will be able to pop in and out” of the Compact does not reflect real-world state legislative operations.

After the Compact is used in one presidential election, additional states are likely to adopt it.

Under the current state-by-state winner-take-all method of awarding electoral votes, general-election campaigns ignore three-quarters or more of the states (as detailed in section 1.2).

In the first national popular vote for President under the National Popular Vote Compact, presidential candidates will necessarily have to solicit votes from all Americans (chapter 8).

Thus, once a national popular vote for President is conducted, many states that did not belong to the Compact in that first election would likely enact the Compact in order to ensure that they would continue to receive attention in the future.

Thus, the Compact is likely to be law in states with considerably more than 270 electoral votes as its second presidential election approaches. Thus, a considerable number of states would have to want to repeal the Compact into order to return to the old system.

Having said that, the nature of democracy is that laws can be repealed. If the Compact does not have sufficient public support, it would simply not govern future presidential elections.

The Compact’s July 20 deadline makes it less vulnerable than the current system to politically motivated last-minute changes by states.

Current federal law (the Electoral Count Reform Act of 2022) provides (in section 1):

“The electors of President and Vice President shall be appointed, in each State, on election day, in accordance with the laws of the State enacted prior to election day.” [Emphasis added]

In other words, in the time up to and including the Monday before Election Day, a state may switch to, or from, the congressional-district method of awarding electoral votes (such as used in Maine and Nebraska), the winner-take-all method, legislative appointment of presidential electors, the whole-number proportional method, or any other method.

In contrast, the National Popular Vote Compact only governs a presidential election if it is in effect in states possessing a majority of the electoral votes on July 20 of a presidential election year.

If anyone is concerned about the hypothetical scenario in which states “pop in and out of” the National Popular Vote Compact, the fact is that the Compact is distinctly superior to the current system in this respect.

The events in Nebraska in 2024 illustrate the difference between the current system and the Compact.

The Republican presidential nominee has won the statewide vote in Nebraska in every election since 1968.

In 1992, Nebraska switched from the statewide winner-take-all method for awarding electoral votes to the congressional-district system.

In 2008 and 2020, the Democratic presidential nominee won one electoral vote from Nebraska’s 2^nd congressional district (the Omaha area). Meanwhile, the Republican presidential nominee won the state’s other four electoral votes.

The Nebraska legislature has considered proposals to repeal the congressional-district method in every year since 2008.

As the 2024 presidential election approached, Nebraska’s Republican Governor Jim Pillen, Donald Trump, and others recognized that there was a politically plausible combination of states that might enable the 2^nd district’s electoral vote to determine the national outcome of the 2024 presidential election. That combination of states (shown by the map in figure 1.22 section 1.6.4) would produce a 269–269 tie in the Electoral College and throw the presidential election into the U.S. House.

When a presidential election is thrown into the House, each state has one vote. Based on the likely composition of the state delegations on January 6, 2025 (discussed in section 1.6.4), the Republican presidential nominee would be chosen President in event of a 269–269 tie in the Electoral College.

On the other hand, if the 2^nd district were to vote for the Democratic presidential nominee in November 2024, and if Nebraska were to retain its current congressional-district method of awarding electoral votes, the Democratic presidential nominee would win the Electoral College by a 270–268 margin.

In the spring of 2024, Pillen, Trump, and others urged the Nebraska legislature to repeal the state’s 1992 law establishing the congressional-district method and replace it with the winner-take-all method of awarding electoral votes.

In reaction to a possible change in Nebraska’s law, Democratic legislative leaders in Maine (the only other state using the congressional-district method of awarding electoral votes) threatened to switch Maine to the winner-take-all method.[987] That countermove by Maine would negate the partisan advantage that would be created by the proposed legislation in Nebraska.

The winner-take-all bill was blocked by a filibuster in the Nebraska legislature. Shortly thereafter, the regular sessions of both the Nebraska and Maine legislatures adjourned.[988]

The Nebraska Governor has the power to call a special session of the legislature at any time. In fact, the Governor called a special session of the legislature starting on July 25, 2024, to consider a limited list of topics largely focused on property taxes.[989]

Although legislation to change Nebraska’s method of awarding electoral votes was not on the agenda of the July 2024 special session, the Governor indicated that he was willing to call a special session before Election Day on that topic if there were enough votes in the legislature to overcome the filibuster, pass the bill, and give the bill immediate effect.[990] Both actions would require a two-thirds vote in the legislature.

Immediate effect would be critical in order to impact the November 2024 presidential election, because newly passed legislation in Nebraska ordinarily takes effect after a 90-day delay.

In Maine, there is a similar 90-day delay before new laws take effect. As in Nebraska, new laws in Maine can be given immediate effect by a two-thirds vote in the legislature. However, because the Democrats do not have two-thirds of the Maine legislature, any attempt by Maine to counter a potential change in Nebraska would have to be enacted by early August.

On August 10, 2024, the Nebraska Examiner reported:

“Nebraska Republican Party Chairman Eric Underwood confirmed what state senators have told the Examiner privately, that the issue is not dead for 2024, and Pillen and legislative Republicans are waiting for the right moment to bring it forward.”

“State lawmakers, including the senator who shepherded the idea last session, State Sen. Loren Lippincott of Central City, have said Pillen would call another special session if he can show the governor he has 33 votes to overcome a promised filibuster.”[991] [Emphasis added]

As of the time of this writing, it is not known whether Nebraska law will be changed in time to impact the November 2024 presidential election.

In summary, a state can change its method of awarding electoral votes right up to the day before Election Day under the current system, whereas under the National Popular Vote Compact, a change in method would have to be enacted and take effect by July 20.

If anyone is concerned about the hypothetical scenario in which states “pop in and out of” the National Popular Vote Compact, the Compact’s July 20 deadline is superior to the current system’s deadline.

The myth that states could “pop in and out of” the National Popular Vote Compact is one of many examples in this book of a criticism aimed at the Compact where the Compact is superior to the current system.

Footnotes