The outcome of the next American elections is highly uncertain and not solely binary (Trump or Clinton). The election results in the United States House of Representatives and the Senate are likely to be also very significant for the governance of the country over the next two to four years. Not only, is there a high probability that the control of the White House and Congress will remain split between Republicans and Democrats after November 8th (less than two month away), but Congress is likely to oppose the newly elected President on several issues even if the President stems from the same party which will control the legislature.
For instance, if Democrats were to take control of the Senate and the House, it is less than certain that a President Clinton would be able to pursue the free-trade agenda and moderate economic policies of her predecessor, Barack Obama. To please the left-wing of her party, Clinton may have to revisit free-trade agreements already in place, reject the Trans-Pacific Partnership proposed by President Obama (in the event that it would pass before the end of his presidency, she might try renegotiating its terms), raise taxes on the upper middle class and more strongly regulate private businesses. These positions would probably go beyond what Hillary Clinton personally considers economically prudent but she may have very little political leeway in the matter.
More importantly, such a liberal agenda could realistically provoke a backlash down the road among independent voters, preventing her party from performing well in the 2018 mid-term elections. The risk of giving back whatever gains she is now likely to make next November in Congress may be substantial; not unlike what happened to President Obama in the 2010 mid-terms. Eventually, such an outcome would further increase political gridlock and could thus also seriously jeopardise Hillary Clinton’s chances of securing a second term in office in 2020.
Similarly, a Trump president insisting on building a wall on the Mexican border, rounding up illegal immigrants, preventing Muslims from entering the country, scaling back the role of the U.S. military on the global stage, restraining globalisation in a vain attempt to bring back jobs to America as well as running large fiscal deficits on the back of populist policies might run into the fierce opposition of moderate leader Paul Ryan’s somewhat more socially inclusive and globalist, yet fiscally conservative, agenda in the House.
Yet, one could argue that such outcomes would still be less likely to provoke gridlock than a Congress controlled by a party ideologically opposed to the President. And there are several valid reasons to argue that, after years of misalignment between the White House and Congress, a gridlock-weary electorate might actually prefer a Congress controlled by the President’s party.
Indeed, the electorate is extremely tired of the status quo: The federal government legislative agenda has been in a permanent logjam in terms of the proposed policies that actually end up being adopted by their leaders. The current situation (a democratic President and a Republican Congress: 54 Republican seats to the 46 held by Democrats in the Senate and 246 Republican seats to 188 for the Democrats - and one vacancy - in the House) is highly frustrating to the American public which perceives partisanship in the government as preventing politicians from pushing forward any sort of constructive agenda.
Given that the approval rating of the currently Republican-controlled Congress is lingering at around 10%, a Democratic landslide would appear to be a desirable outcome by an American electorate eager to see some resolution. A unified government would normally be seen as helpful at reducing frictions and allowing a productive legislative agenda, whatever it may be, to be implemented as it would help bills move forward within the two Chambers and reach the president without risking a veto to finally become law.
Moreover, more than ever, the U.S. electoral process seems to thrive on partisanship. A consequence of this heightened political polarisation is that voters are more inclined to support an entire slate of candidates from the same party rather than pick and choose candidates from different parties according to their skills, experience and local economic, political and social agenda.
And indeed, over the recent decades, the voting pattern within each American congressional district has increasingly followed party lines. Twenty-four years ago, a significant proportion of the electorate backed different parties when voting for their preferred presidential candidate and local representatives in Congress. Fast forward to the 2012 presidential elections and less than 5% of congressional districts in which voters supported a presidential candidate from one party were seeing representatives from another party win the popular vote in the district. This voting pattern would certainly support the election of a Democratic Congress in the case of a Clinton presidential victory and a Republican majority if Donald Trump were to be elected U.S. president.
However, there are even stronger arguments supporting the view that Democrats and Republicans will again end up sharing political power, thus maintaining an unproductive political atmosphere and prolonging the status quo in the conduct of government affairs. Even if according to recent polls, Hillary Clinton is favoured to win the presidency, she is not popular; only 33% of voters view her positively; Donald Trump performs even worse with a score of 24%. In order to ease the political stalemate in Washington, independent voters may have been inclined to support a Democratic Congressman had Barack Obama - who still enjoys a 50% approval rating - remained President. In this election, many independent voters may end up voting for Hillary Clinton only to avoid a Trump presidency. But, they are also likely to support their local Republican Congressman to express their distrust of Democratic presidential candidate Hillary Clinton and/or their disagreement with her increasingly liberal agenda.
Moreover, although still within reach, winning 5 new seats to reach a majority of 51 seats in the Senate will be a stretch for Democrats: Of the 34 Senate seats that are up for grab this year, 24 are defended by Republicans incumbents who currently leading in most of the toss up states according to recent polls. But more significantly, winning the House of Representatives, where all 435 seats are up for re-election next November, remains a very long shot. Even if Democrats were to pick up 5 to 15 seats from Republicans in November (a wide range consistent with recent polls), the GOP would still maintain a comfortable majority in the House.
Besides, neither of the presidential candidates is truly appealing to many of their own party members and independents: only 2% of the electorate has a favourable view of both candidates; a number which is much lower than in previous presidential elections. At both ends of the political spectrum, partisanship and party line may still characterize the voting pattern and favour alignment. However, among independents and more centrist voters within both parties, a power split between the Presidency and Congress may represent an insurance policy against erratic behaviour from either presidential candidate or Congress.
For instance, the Never-Trump movement among independents and within the Republican Party is gaining strength. However, its supporters, regardless of who they will ultimately vote for in the presidential election (Clinton, Johnson from the Libertarian or simply not vote), are highly likely to oppose the Democratic candidates in their own congressional district for fear that a Democratic congress would unleash an extremely liberal agenda if Clinton were to become president.
Although for a majority of Canadians, according to our own unscientific evaluation, a Clinton presidency and a Democratic Congress would certainly appear to be the most logical outcome to resolve the current gridlock, the American electorate is to the right of the median Canadian voter. As such, a large proportion of Americans, even among independents, at least at this point in time, probably still perceive the current anti-business and environment-friendly Democratic agenda of high taxes and strict regulations to be detrimental to the U.S. economy. Moreover, if only on social issues such as gun control, abortion, the death penalty, gay rights and Supreme court nominations, it is very difficult to see Never-Trump Republicans supporting Democratic candidates running for Congress.
In the case of a Clinton likely victory, Republican and independent voters - who dislike Hillary Clinton but who would vote for her anyways to avoid a Trump presidency - would thus be more inclined to vote for Republican candidates in their congressional districts.
Conversely, faced with the inevitability of a Trump presidency, many independents voters who dislike Hillary Clinton may choose to vote for the Democratic representative in their congressional district. The strategic votes would be cast to counter the unpredictable behaviour they think a President Trump could exhibit as President if they extrapolate from their recent experience of the candidate’s conduct during the present campaign. However, as mentioned above, the hurdles for Democrats to grab power in both the Senate and the House are high as they need to win a lot more of the contested senate seats and congressional districts to gain a majority in the two Chambers.
In our opinion, because Republicans already control the House and the Senate, even if a large proportion of conservative and independent voters dreading a Trump victory, decided to vote for the Democratic candidates in their congressional districts, it would almost certainly be insufficient to hand Congress over to the Democratic Party.
Finally, if they start feeling that the chances of a Donald Trump victory are slipping, the leaders of the Republican Party may push their congressional candidates to distance themselves from Trump’s ideas and rhetoric. Distancing themselves from “outsider” Donald Trump, would be relatively straightforward for Republican candidates wanting to attract independent voters unhappy with Ms. Clinton and Never-Trump Republicans. The tactic for candidates would be to stick to a conservative agenda in order to try increasing their odds of keeping control of Congress, where their chances of winning are greater than those of Democrats. Even though this would further diminish M. Trump’s odds of winning the presidential race, this fall-back strategy would at least reduce to a minimum the chances that the Democratic Party could gain control of Congress.
Democratic candidates, on the other hand, may have more difficulties distancing themselves from “insider” Hillary Clinton for fear of reducing her chances of winning the presidency. Without great odds of winning a majority in both the Senate and the House, they could be left empty-handed without the presidency, the Senate, nor the House.
Therefore, it is our view that, in this election at least, because Republicans candidates are naturally less aligned with the positions of their own presidential candidate than Democratic candidates are with Hillary Clinton and because Republicans already control Congress, the particular dynamics of this election campaign favour Republicans in Congressional contests.
Figuring out the outcome of the presidential election thus amounts, more than in any previous presidential elections, to solving a game theory problem in which the reaction of voters depends on their perceived intentions of how other electors might vote.
In the end, given Clinton’s current lead in the poll, here are our estimates of the probabilities of the different outcomes in the November elections. These probabilities are not poll results but they are based on various polls and our own analysis. If, for instance, polls were to change and Clinton’s current 5% lead in the polls were to fall to below 2%, these probabilities would change and reflect that Donald Trump’s chances of winning the presidency would become higher than the 25% odds we attributed to this outcome in the table below. In fact, as we were putting the final touch to this document, on September 9th, an L.A. Times / USC tracking pool reduced Clinton’s lead over Trump to a mere 1%.
According to our estimates - obtained by adding up the probabilities from the above table - the probability of a power split between the Republicans and the Democrats is 85%. Moreover, it is worth noting that, even if winning the Senate could be within reach for the Democratic Party, winning a supermajority is not in the cards. The probability of a unified Democratic government is 5% and that of a Republican one is 10%.
Economic and market implications
Sticking with the most probable scenario given current polls, market implications are relatively straightforward compared to other scenarios. Unless anti-free trade Republicans gather allies in the left wing of the Democratic Party and threaten to undo existing trade agreements, the status quo will likely prevail and somewhat please market participants who fear most the uncertainty of alternative scenarios. The probability of meaningful fiscal reforms and higher taxes would also remain very low and the legislative deadlock would persist.
On the other hand, the negative consequences of a gridlocked government would likely be felt in the longer term. For instance, the fiscal package several policy-makers are hoping for in order to lift the world economy out of its current liquidity trap - where low interest rates fail to stimulate investment - would unlikely pass Congress or would be expected to be underwhelming. Even though the platform of both candidates calls for a higher deficit, with a divided government, the lower corporate and individual tax rates and cost-cutting measures proposed by Republicans would probably be irreconcilable with the Democrats’ desire to increase taxes and spending.
With luck, the next mid-term elections would bring some resolution to the lasting political impasse and allow more constructive policy proposals to move forward. In the meantime, investors would probably have to wait and hope that they are correct in expecting that the most damaging proposals currently being discussed by the candidates won’t see the light of day.
For Canada, continued American political stagnation would be the most favourable outcome: Despite strong opposition from both Democrats and Republicans, NAFTA would probably remain in place as is; and our economy, via its exports, would continue benefitting from the U.S. economy which will most likely remain on its current tepid, yet positive, growth path. In this scenario, commodity prices, including oil, would most probably rebound and favour Canada’s recovery. Equity markets would benefit as well as the Canadian dollar. Eventually, interest rates would gradually increase to reflect falling unemployment and rising inflation. And although monetary policy would tighten in the U.S. before it does in Canada, the Bank of Canada would be expected to follow the Fed, at least partially, shortly after.
All other scenarios carry the risk of jeopardising our trading relationships with the U.S. In the worst case scenario, the November 2016 election outcome would most likely provoke an American recession, triggering along the way a global recession which Canada would be one of the most obvious and immediate victims.
A step back for free-trade would be especially damaging for Canada (our merchandise exports to the U.S. totalled 400B$ or 20% of nominal GDP last year). Not only is the country counting on exports to reinvigorate its anemic economy but a retrenchment in globalisation would also affect the demand for its commodities, especially oil. In such a situation, Canada would see the value of its oil exports fall further. The volume of its non-commodity exports would most certainly plunge as well. Interest rates would remain near zero and could become negative and the Canadian dollar would reach new lows.
For now, polls are favouring Hillary Clinton and a Republican Congress and markets are still downplaying a Trump victory. However, there is plenty of time left before the election for various events to unfold and affect the probable election outcome. This means that, before November 8th, there could still be several reversals in the perceptions of market participants regarding the likely victors of the November elections. In our view, the current period of calm offers the opportunity to build a defensive position in case the election results end up displeasing markets for any reason: A Trump presidency, a Congress unfavourable to free-trade, credible announcements of the federal government’s intentions to increase regulations and/or raise taxes, failure to deliver economic growth, etc.
Any reversal in current sentiment would offer investors, who would have already built a position against such risks, an opportunity to take some profits and maintain some protection at a lower cost. By repeating the strategy, as probabilities of an unfavourable outcome (and resulting volatility) fluctuate, it could be shrewd to try to continue taking advantage of the instability and gradually build a larger defensive position using a minimum of resources. This would insure one’s portfolio against a disastrous watershedevents, like the election of Donald Trump as President in November or increased expectations of a globalisation debacle next year.
Moreover, to further reduce the cost of such a strategy, it could be attractive, while U.S. equity markets are near-record levels and the Canadian dollar trades at a decent level, to use collar positions to simultaneously protect against a harmful election result while giving up some of the upside; especially if one believes that the upside appears to have limited potential and a low probability of surpassing already heightened expectations of favourable growth. Again, any correction would offer investors the opportunity to partially unwind such positions at a profit to maintain some protection at a more reasonable cost.
In terms of foreign exchange, the dynamics of an unfriendly election outcome on the value of the US$ is not easy to predict. At first glance, U.S. protectionist policies would appear to favour the Yen and the Euro as Japanese and European businesses could take advantage of the situation to extent trading relationships in Asia, in particular with China, at the expense of U.S. companies. Yet, the retreat of the U.S. economy from the globalisation process would likely be felt beyond its borders and provoke significant growth setbacks in Asia and Europe as well. Capital would then be expected to seek the safety of the U.S. dollar. We would thus anticipate the greenback to outperform other major currencies but its outperformance to be limited by its deteriorating current account. On the other hand, the Canadian dollar and other commodity currencies would surely be the most heavily hit.
On the bond market, because of their status as safe haven assets, U.S. bonds would likely perform well in case of an unfavourable election outcome; while credit spreads would get much wider to account for the increased risks. The Fed would most certainly also abandon plans to raise rates and potentially engaged in a new round of quantitative easing or other unconventional measures. The end result would probably be a flattening of the yield curve as market participants would postpone indefinitely an eventual interest rate hike by the Fed.
Moreover, if Donald Trump’s proposal of offering U.S. companies a tax holiday (from 35% to 10%) on the repatriation of foreign profits were to gain support in Congress, it would tend to lower U.S. interest rates further and cushion the widening of corporate spreads. It would also lead to an appreciation of the U.S. dollar and boost the share price of the firms able and willing to take advantage of the tax break. However, we do not believe that it would bring significantly more business investments in the U.S. economy.
In Canada, in the anti-globalisation scenario, as mentioned above, the Bank of Canada would probably move swiftly to lower short term rates to zero and even into negative territory. Yet, Canada’s sovereign credit could likely be affected negatively. As a result, the Canadian yield curve could steepen; if it did flatten to account for “lower-for-longer” policy rates, it would do so less than the U.S yield curve. The negative interest rate differentials between Canada and the U.S. on the longer horizon of the yield curve could thus shrink and probably reverse back into positive territory where they historically have been, most of the time, during the quarter-century prior to the last financial crisis (see graph below).
