Dangerous Electoral Tinkering and the Law of Unintended Consequences

The issue comes up after virtually every Presidential election - is there a need for a change in the way our Presidential election process works?

Following the 1992 and 1996 elections, Nebraska and Maine, frustrated that no one paid attention to them in Presidential contests, with Nebraska beings solidly red and Maine being solidly blue, passed laws that went into effect for the 2000 election that split their votes based on congressional district, with the winner of each congressional district receiving 1 electoral vote and the winner of the statewide vote receiving an additional 2 votes.  Nebraska and Maine are small, meaning the overall electoral vote isn't that significant.  In fact, Maine has never actually split its vote, since in every election from 2000 to 2012, both of its congressional districts, along with the state vote, went to Democrats.  Nebraska has split its vote only once, with Barack Obama winning 1 congressional district in 2008 and giving himself 1 of Nebraska's 5 electoral votes in a vote that hardly mattered given his large electoral margin over John McCain.

The issue of the electoral college certainly came up in the aftermath of the 2000 election, which provoked many to push for a system where the winner of the national vote won the election.  The National Popular Vote Interstate Compact, which I've written about before, was a creative system which would give the winner of the national vote the majority of electoral votes without the need for a constitutional amendment to that effect, gained steam in 2007, being enacted into law in 9 states over the time period from 2007 to 2011, giving it 49% of the electoral votes that it would need to take effect (approval in California in 2011 was a big contributor to the electoral count.)

In 2012, Republicans in Pennsylvania briefly toyed with a system similar to Nebraska and Maine's, which would have been very significant given the size of Pennsylvania and how favorable its current congressional districts are to the GOP, but abandoned the idea after it provoked a strong public reaction against it.

Now, other GOP-controlled states that voted for President Obama are taking up the issue.  The legislature in Virginia is considering a proposal, called "intriguing" by RNC Chair Reince Preibus, that would allocate 1 vote for each congressional district and 2 votes to the winner of the most congressional districts.  This scheme would have effectively given Mitt Romney 9 of Virginia's 13 electoral votes, despite losing the popular vote in the state, because of the concentration of Democrats into relatively few congressional districts in the state.

Before I discuss the merits of various proposals around the electoral college, it is probably beneficial to understand the history that brought us to where we are today.

The intention at the time of the writing of the constitution was never for the people to have a direct say in the election of the President.  The electoral college was created with the intention of it being a learned group of men (yes, they were certainly all men in those days), appointed by state legislatures, charged with finding the best candidate to be President and Vice President.  In fact, the original text of the constitution didn't even specifically charge them with casting a vote for Vice President, the Vice President was just to be the runner-up in the Presidential race, with each elector voting for two candidate for President.

This system worked fine in the early years - George Washington was a unanimous choice in 1788 and 1792.  In 1796, the first post-Washington election, the true formation of political parties began in the US and regional and partisan divides began to occur.  As the more "states rights" Democratic-Republicans controlled the South, Democratic-Republican favorite Thomas Jefferson was elected (the South had a large electoral advantage in those days) and Federalist John Adams carried the North.  The issue of not picking separately for President and Vice President was quickly exposed, with Jefferson and running mate Aaron Burr receiving identical numbers of votes for President from the Democratic-Republican electors, which thrust the race into being decided by the House of Representatives.  The 12th amendment was then ratified prior to the 1804 elections to rectify this issue and specify separate ballots for President and Vice President.

Also in 1804, the system for choosing electors, which had been left up to the states, began to evolve.  6 states had moved to the system that we generally know today, of holding a statewide popular vote and giving all electors to that candidate.  4 states had adopted the congressional district-splitting plan that some Republicans are now advocating.  1 state had a mix of state legislature-chosen electors and popular vote electors.  And finally, 6 states still had the state legislature picking electors.

This mixed method did not really expose itself as a flaw as Thomas Jefferson was a wildly popular incumbent and won election easily, carrying 162 of 176 electoral votes.

The mixed method of choosing electors continued for some time, with a gradual trend towards statewide popular vote determining electoral college representation.

By 1824, 12 states were choosing electors by statewide popular vote, 5 states were divided into congressional districts, 6 states were choosing by the state legislature and 1 had a mixed model.

By 1832, all states except 2 were using the statewide popular vote method, with 1 state using the congressional district method and 1 state appointing by state legislature.

By 1836, South Carolina was the lone holdout, continuing to appoint by state legislature, with all other states using a statewide popular voting system.  South Carolina held out until after Civil War reconstruction, at which point it joined the other states in a statewide popular voting system.

The statewide popular voting system then became the standard until 2000, when Maine and Nebraska made their changes.

The lessons that we should take from this history are:
* Rather than being a thought-out constitutional system, our modern process for choosing electors is something that evolved through trial-and-error in the states and in many election cycles.
* Having a few "outlier" states that have different methods of choosing electors did not occur for 150 years, but has quite a lot of history in the early days of the Republic.

All of this brings us to today.  The constitution continues to permit each state to decide how to allocate its electors.  So, what the GOP is proposing is certainly within the constitutional authority of each state.  So, the obvious question, both from a state-level and a national perspective on any changes our existing system is, what are the values and incentives around which we should design a system?

I'll take a shot at this, in the context of our modern political system.

Any system should seek to:
a. Reflect the will of the voters it seeks to represent in the electoral college
b. Provide for the interests of the individual state
c. Produce a clear result in the Presidential race
d. Produce a result that is consistent with a democratic (small d) election
e. Resist subversion of the political process by narrow interests

Our existing system is fairly strong at A, C, D and E. 

The statewide winner wins the electors in 48 states.  That's pretty representative.  Yes, in close states, the 49% who lose may feel unrepresented, but elections have winners and that is sort of part and parcel to losing.

The results are generally very clear in Presidential races.  The 2000 election was strikingly close in Florida, but in the 136 years between the Tilden/Hayes dispute, which was fueled both by being a very close election and by the scars of reconstruction, 2000 is the only election where the outcome was legitimately in doubt for a significant period of time.

Our system is also fairly democratic.  The candidate who won the largest plurality of votes did not win in 2000, although it was very close.  Prior to that, you have to go back to Benjamin Harrison in 1888 to find a candidate who won without winning the poular vote.

Our system also resists subversion.  It is very hard to make a new state and, by constitution, existing states can't be divided without both their consent and that of Congress, making it hard to rig the game.

Where our existing system largely fails is in B.  The interests, money and attention all go to a narrow band of states that are considered swing states.  Broadly, in today's terms, this is Florida, Ohio, Virginia, Iowa, North Carolina, Colorado, Pennsylvania, Nevada, New Hampshire, Wisconsin, Michigan and Indiana.  No Presidential candidate visits our three largest states: California, Texas and New York, except to raise money, because those states don't have outcomes that are in question - California and New York are reliably blue and Texas is reliably red.

The Interstate Voting Compact would do an excellent job in all five categories.  The winner of the national vote would win.  Every state would get attention in proportion to its population...a vote in New York City or Dallas would be just as valuable as a vote in Tampa or Denver.  Results would be clear - yes, this would have likely led to a larger recount effort in 2000, but Al Gore won the national popular vote by 0.5% or over half a million votes, a far more meaningful and less questionable margin than 0.0092% margin in Florida by which the election was decided.

The major criticism of this system would be that it leaves smaller states behind.  The electoral college was intentionally created as House seats + Senate seats in order to give large states more sway than small state, but less sway than their population would dictate, so that the smaller states still had a strong stake in the system.  This is a legitimate concern, but versus in our present system, how many small states really get attention?  Maybe New Hampshire.  But that's about it.

The proposed congressional district system has some history behind it.  As I noted above, it was used by several states for several election cycles in the early days of the Republic.  But it also has major flaws.

It can easily fail the "will of the voters" test as 2012 example from Virginia attests - the candidate receiving less votes receiving over 2/3rds of the electors is a pretty odd thing. 

It would do a good job promoting the interest of individual states, as candidates would have to campaign district-by-district and couldn't take very many electoral votes for granted.

It could be a potential disaster in providing decisive results.  Can you imagine a close race where there are 30 or 40 congressional districts that are all within the margin of a recount?  Multiply Bush vs. Gore by 30 or 40 and that's what you would have.

It would also tend to produce results that are less democratic (and also less Democratic, as of today), potentially leading to somewhat democratically inexplicable results.  Mitt Romney won 225 congressional districts in 2012 and 24 states, would would have given him 273 electoral votes to Barack Obama's 265 if the whole nation had used a system where the congressional district winners won 1 vote and the statewide winner won 2.  This, despite President Obama winning the popular vote by almost 4%.

In short, the GOP idea is a bad idea.  The appeal to them is obvious, from the description of the 2012 race above.  But it is also short-sighted.  A major reason why Romney won so many districts was Republican control of state houses when district lines were redrawn for 2012, following the 2010 census, allowing for gerrymandering of districts to support GOP victory.  This could easily flip in the next census.  Also, since the system is not being broadly adopted, adopting a system like that just for a few states, could actually hurt them in a close election.  What if just Virginia had adopted the system this time around and Romney had won it but left 4 electoral votes for Obama because of the system which proved decisive?

If we are not going to have a national popular vote system (which I favor) because small states don't want to give up authority and we want to reform the system, a better approach would be proportional representation in the electoral college.  Give the winner of the state 2 electoral votes and split the other votes by percentage of the vote.  For instance, in Iowa, which has 7 electoral votes, you'd give 2 votes to the winner and 1 vote each for approximately 20% of the vote garnered.  This would have split Iowa's vote 5-2 for Obama, versus the 7-0 that we saw.  In California, the vote would have split 35-20 rather than all going to Obama.  In Texas, it would have split 23-15 for Romney, instead of all going to Romney.  This would encourage campaigning in all 50 states and would take away the concentration of power from a few swing states.

Of course, such a system would only work if it is broadly adopted.  If Texas does it but California does not, then it just amounts to an unfair 15 electoral vote advantage for the Democrat.

My guess is that despite the discussion, not much will happen on this front for the 2016 election.  Governor Bob McDonnell of Virginia has already stated that he doesn't see an issue with the states present electoral system and cooler heads are likely to prevail in other GOP-controlled legislatures.

But all of this does provoke an interesting question as to whether our current system for electing Presidents is the best one we could find.

If you like this site, tell your friends.

SOTU Recap: Was That Reagan in 1982?

I miss the days when State of the Union speeches typically lasted 40 to 45 minutes. Bill Clinton set the bar for insanely long speeches, stretching as long as 90 minutes some years. And President Barack Obama last night continued that ugly trend, going almost 75 minutes by my count.

Here's my quick recap on the good and the bad:

The Good
1. He was funny, self-deprecating and entertaining for the first half of the speech. I loved the line referring to health care "as should be clear by now, I didn't do it because it was good politics."
2. He stuck to his guns. Health care, still needed. Cap and Trade, get it done.
3. His explanation of the economic circumstances facing the country and his administration's approach was excellent.
4. He FINALLY loudly called for Congress to allow gays to serve openly in the military.

The Bad
1. No path forward - how are we to reconcile the political reality of today with the ambitions that he laid out.
2. False bi-partisanship - the repeated rhetoric of last November, with very little tangible action. We'll buy it when we actually see Democrats and Republicans working together.
3. Small ball, and not even good small ball - small business capital gains taxes? Are you serious? Small businesses aren't hiring because they can't get credit and don't have customers, not because their owner is worried that when he makes millions, he will have to pay a 15% tax. Not a single new big initiative.

In total, it was a decent, but fairly forgettable speech. I expect the President's approval to get a modest bump from the speech. It will do nothing in and of itself, to move congress forward. The President's actions in the coming weeks are critical if he is serious about getting real bi-partisanship or getting health care or cap and trade done. He's going to have to be a lot more directly involved in the process and a lot more active.

So, why the title of this blog? Take a look at the link below to view President Ronald Reagan's 1982 State of the Union Speech.

http://www.youtube.com/watch?v=2QFLsxeEl5I

I watched it today because I was interested in how a different President, from a different party, reacted to similar economic circumstances. Interestingly, while the solutions proposed are different, the construct is remarkable similar. Both talk about the crisis they were handed. Both defended their programs, saying things would have been far worse without their actions. Both expressed optimism, without really committing to timelines or metrics for improvement. Both spent more than half of their speech on economic matters, with only small sections on foreign policy issues, despite the prominence of international concerns in their administrations.

It should be noted that Reagan's party went on to get smacked in November 1982, but that Reagan won in a landslide in 1984. We'll know in a few months whether the first half of that repeats for President Obama.

A few quick thoughts on Bob McDonnell's GOP response:
It was a very good response speech. Having a live audience worked much better than a single individual staring at a camera. And Gov. McDonnell (R-VA) is clearly a gifted communicator. The speech was respectful and upbeat, while drawing stark policy contrasts.

In short, it was much, much better than Gov. Bobby Jindal's (R-LA) awful response speech a year ago. While it would be early to suggest that a star was born last night, it is clear to me that McDonnell has the potential for higher office.

If you like this site, tell your friends.

2009 Update, Is Obama at Turning the Corner?, Mixed Economic Signals

2009 -- Still Looking Good for the GOP, But Small Reasons to Hope for the DEMS
Will the GOP sweep the New Jersey and Virginia governor races in 2009? It still looks likely, but perhaps slightly less likely than it did a few weeks ago -- especially in New Jersey.

Let's begin in Virginia. McDonnell still leads Deeds by 8 to 15%, depending on which poll you believe. Deeds has scarcely led at all since his come-from-behind win over the much better known Terry McAullife, leading only one poll (an early June Rasmussen poll.) The Real Clear Politics average shows McDonnell at +9.8%, my own sample-weighted methodology puts it at +9.0%. Either way, this one still seems to be a likely GOP pick-up, as Virginia does not historically turn left at the last minute.

In New Jersey, unpopular Gov. Jon Corzine (D) still trails upstart conservative Republican Chris Christie by 5-10% depending on your poll. The RCP average shows Christie +6.5%, my own methodology has Christie doing even better at +7.9%. But this actually shows a significant tightening, as Christie led by double digits throughout all of July and the first half of August. I see a familiar pattern that has played out many times in New Jersey over the past 20 years -- tax-angry voters initially favor the Republican outsider, then, on closer reflection, swing back to their natural blue tendencies. I'll rate this as lean GOP pick-up for now, but I wouldn't be surprised at all if I were calling this a pick 'em by October and projecting a close Corzine win by November.

All-in-all, the GOP has nothing but upside (they are fighting for two seats currently held by Democrats) and still seem poised to take at least one and possibly both of them. But the possibility of massive routs that would have an impact on national policy seems to be waning.

Stabilizing Obama
After almost two months of solid free-fall that cut Obama's approve minus disapprove from +28.3% on June 27th to a low of +8.7% on August 30th, the President's scores have been spiking int he past few days and sit at +11.7% as of September 3rd, a full 3 point improvement in 4 days. This has largely been driven by significant improvements in the Gallup (+9% in 4 days) and Rasmussen (+7% in 4 days) tracking polls, which show wildly divergent overall results (Rasmussen has Obama at -1% today, Gallup at +17%), but a similar trend. Whether this is just a blip, a stabilization or the beginning of a recovery in Obama's numbers remains to be seen. But he still hasn't crossed the 7.2% margin of his November victory...yet.


In his monthly totals, you can see just how big a toll the summer took on Obama's numbers. In 2 months, his monthly averages (which flatten out the spikes and dips) dropped by 14%. September has started slightly lower than August, but for the first time in a long time, the President's daily numbers are ahead of his monthly numbers, which would typically project an increase in his monthly numbers, at least in the short term.

Can We Be Recovering with Unemployment Spiking?
The BLS today announced that the unemployment rate in July had surged by 0.3% to (another) 26-year high of 9.7%. Include underemployed and workers who have given up and you have 16.8%.

So how can I be saying we are in recovery?

As I've said before, unemployment languishes for long after a recession ends (and make no mistake, I believe this one ended this summer.) It is likely the unemployment rate will reach low double digits before it starts declining, towards the end of the year. And the decline may be slow. We "only" lost a little over two hundred thousand jobs in July, which would typically equate to a 0.1% bump in unemployment. The rate went up more, because more of those "given up" workers who don't count in the official tally, have no re-entered the work force. Expect more of the same. But note the stock market actually rallied today on the news, which was slightly better than expected.

Protracted unemployment still poses a huge political issue for the President. The explanation above is not something that President Obama can easily give in a way that is accepted by people. As long as people are out of work and hurting, they won't feel the recession is over.

So, let's get moving on this stimulus!

Latest numbers from recovery.gov:
Estimated Tax Cuts Paid Out: $62.5B (21.7%)
Spending Authorized: $217.0B (43.5%)
Spending Completed: $88.8B (17.8%)

$4.2 billion in stimulus spending happened last week, far above the average since the bill was passed. Still, at that pace, it would take 98 more weeks, almost two years at that pace, to complete spending. That won't be acceptable unless unemployment falls.

Inflation, Pfft
Among the dizzying theories about the end of the recession (is it V-shaped? Or W-shaped? or U-shaped? Or square-root shaped?), one that I don't buy at all is the theory that all this government spending and loose monetary policy will lead to massive inflation.

Inflation occurs when demand significantly outstrips supply of something. With unsold business inventories, house prices way off peak, oil at under $70/barrel and factories way under capacity, I can't see what that thing will be, at least for the next several years.

Sure, the Fed has to be careful to tap the brakes when the time comes...but that time is probably 18 months from now. Inflation risk is very small in the near-term.

Clearly the investing public things so too. The 10-year bond is yielding under 3.5% and the spread between 10-year and 30-year bonds is only 0.7%, implying very little investor fear about inflation, even over longer time horizons. TIPS, which pay a fixed rate over inflation, are yielding 1.8% plus CPI on a 10-year bond, meaning that the investing community has pegged 10-year inflation at a 1.7% rate (the spread between Treasuries and TIPS, which contain all the same risk-aspects, save for the inflation factor.) Hardly the late 70s and early 80s. Inflation may come out higher than that, but inflation under 5% is generally not an inhibitor of economic growth. And I'm not worried that the market is THAT wrong.

If you like this site, tell your friends.