A Contrarian Take on the Trade War

Last year, the United States imposed a 10% tariff on $200B worth of Chinese goods with a 25% tariff scheduled to take effect at the end of 2018 [1]. In December, President Trump and President Xi Jinping announced a temporary reprieve, postponing further escalations for another 90 days [2]. By all accounts, both sides need a deal. China’s economy is slowing down much faster than expected, and there are growing concerns about the impact of the government shutdown. Both sides show willingness to deal, with China’s Vice Premier scheduled to meet with US Trade Rep Lightzhier and Treasury Secretary Mnuchin later this month [3].

Despite public overtures, Chinese concessions, and a need on both sides, we should be skeptical. The wildcard remains, and has always been, President Trump. He needs a win, badly. 51% of the American public blames him for the shutdown [4]. His approval ratings, though declining, remain stuck between 40%-44%. On the surface, one might think this statistic is meaningless. Trump won the 2016 election with historic disapproval numbers and he doesn’t need much more than ~44% to win re-election due to his strengths in the electoral college. However, there are emerging signs of weakness. White voters without a college degree, a core constituency also concentrated in swing states, are at their lowest levels. His approval rating among this group has fallen 6% [5].

Taking a tough stance on China remains one of the few ways to regain the support of this demographic. Anti-trade, anti-China, and anti-Mexico rhetoric helped Trump win these voters, and it’s likely that he’ll use it again it to win them back. The implication is worrying: Trump is willing to throw negotiations with the Chinese to look tough, save face, and show that he’s standing up for one of his campaign promises. If he meets the Chinese halfway, he scores economic and political wins but not the wins he needs to regain the support of his base. Trump would rather have a 42% approval rating and 90% with his base than a 45% approval rating and less support with the base. The base shows up to vote.

If Trump pursues this approach, financial markets will plummet. It’ll erode one of Trump’s core messages for re-election: “The stock market is doing well because of me.” At the end of the day, none of that matters for Trump. He cares about winning, no matter what the cost. Unfortunately, he’s running out of options. Taking a hardline stance on China is one of the few moves left.


[1] https://www.cnbc.com/2018/09/17/trump-puts-new-tariffs-on-china-as-trade-war-escalates.html

[2] https://www.bloomberg.com/news/articles/2018-12-13/white-house-said-to-officially-delay-china-tariff-hike-to-march

[3] https://www.politico.com/story/2019/01/15/liu-he-washington-trade-talks-1084246


[4] https://www.reuters.com/article/us-usa-shutdown-poll/half-of-americans-fault-trump-for-shutdown-despite-his-blaming-democrats-poll-idUSKCN1P92QV

[5] https://www.cnbc.com/2018/09/17/trump-puts-new-tariffs-on-china-as-trade-war-escalates.html

Leaking the Mueller Interview Questions Helps Trump (for now)

Yesterday, the New York Times published the list of questions Bob Mueller wants to ask Trump. They range in style and substance. Some focus on Trump’s intent, a key component of obstruction of justice charges, whereas others focus on what Trump knew at specific points during the campaign, like whether Trump knew about campaign officials’ efforts to get Russian help during the campaign. The pundits, politicos, and legal experts have picked apart the questions themselves, but I’m more interested in the leak.

I think someone from the Trump team leaked it, which helps Trump and the Trump-o-sphere (Fox News, Devin Nunes, etc.) in two ways.

First, it underscores the severity of the investigation. Trump is well known to embellish, obfuscate, and generally not tell the truth. Questions like “What was the purpose of your February 14, 2017 meeting with James Comey, and what was said” or “Why did you hold Mr. Sessions resignation until May 31, 2017, and with whom did you discuss it” are likely to cause problems. If Trump lies, that’s a crime and one that Republicans in Congress feel very strongly about. Keep in mind that Bill Clinton wasn’t impeached for philandering. He was impeached for lying.

Second, leaking the list of questions helps Trump’s legal team make the case that they can submit the answers in writing. Per the New York Times, they already tried that and came to this list as a compromise for an in-person interview. But this helps them make the same case again, and in public. It’s easy to ignore the fact that the questions are meant to serve as a starting point for discussion. Mueller and team can dig into whatever direction they please.

The biggest threat to the Trump presidency is either A) an emboldened Democratic House and Senate that moves forward with impeachment and/or B) a face-to-face interview with quite possibly the best legal team in the country. Neither look good, but the Trump administration arguably has more control over the latter. Leaking the questions and having the media make the same case that Trump’s lawyers have been making to their client isn’t a bad strategy for our cable-television-obsessed-president. It might just save him for the time being.

tl;dr Pardons

A few days ago, the Washington Post broke a story that Trump and his legal team were looking into pardons, including Trump’s family members and even himself. According to Article 2, Section 2 of the U.S. Constitution —

The President shall be Commander in Chief of the Army and Navy of the United States, and of the Militia of the several States, when called into the actual Service of the United States; he may require the Opinion, in writing, of the principal Officer in each of the executive Departments, upon any Subject relating to the Duties of their respective Offices, and he shall have Power to grant Reprieves and Pardons for Offenses against the United States, except in Cases of Impeachment.”

The constitution is quite clear. The president has the authority to pardon anyone for offenses against the United States (read: federal crimes). But, the question about whether or not he can pardon himself is up for debate. That’s the “except in Cases of Impeachment” part.

Most legal experts, including Justice Department lawyers in 1975, agree that the President cannot pardon himself. According to a Justice Department memorandum written during the Ford administration, “Under the fundamental rule that no one may be a judge in his own case, the President may not pardon himself.” Obviously, this hasn’t been tested. Nor did the framers consider this particular example.

Let’s assume that Trump does pardon everyone around him for any potential crimes. While this may kill Bob Mueller’s investigation into any connections to Russia, it may bolster his case that Trump continues to obstruct justice. Obstruction of justice is anything that impedes a current investigation – something that Trump may have done when he fired Jim Comey. Pardoning family members to prevent Mueller from further looking into them may be a textbook example of obstruction of justice as well.

Even if Mueller has a solid case against Trump, either for obstruction or collusion, it’s hard to indict a sitting president. According to the New York Times, the Justice Department, in 1973, concluded that “structural principles” shield a president from immunity. They said that “the stigma of being indicted” would interfere with the duties of the executive branch and prevent it “from accomplishing its constitutional functions”. Kenneth Starr, the special counsel who investigated Bill Clinton, looked into this much more extensively. His office concluded that presidents can, in fact, be impeached — “…if he cannot be prosecuted for violating criminal laws, he will be above the law.” Recall that one of the main criticisms of the founders was that Kings were effectively above the law with nearly unlimited power. Starr’s office circumvented the “structural principles” objection by pointing to the 25h amendment, which formally established a line of succession for the presidency. The executive branch can continue to function even if the president is indicted; the vice president will takeover.

It’s hard to tell where Mueller and Trump, especially Trump, will go next. Needless to say, this has not happened before. If Trump pardons everyone for “the Russia thing”, whether or not he includes himself, there will certainly be a political, legal, and constitutional firestorm.

Why did the AHCA Fail

On Friday afternoon, minutes before the House of Representatives was set to vote on the AHCA (Obamacare replacement), Speaker Paul Ryan yanked the bill. He didn’t have the votes, and the last minute negotiations between the House leadership, President Trump, and the Freedom Caucus stalled. It turns out that it takes more than 7 years, dozens of repeal votes in the House, multiple national elections, and full control of the White House and Congress to scale back one of the largest improvements to our social safety net since arguably the 1960s.

The bill failed because Paul Ryan never had the votes to begin with. There are 237 Republicans and 195 Democrats in the House. From the onset, it was clear that 0 Democrats would vote for the bill, including those in conservative districts that Trump handily won. After the CBO released their estimates (24 million to lose coverage within 10 years), moderate Republicans began to walk back their support. Don’t even get started on the Senate. The original (and amended versions) of the AHCA were non-starters. The bill was rushed through committee before the CBO analysis (nonpartisan economic scorekeeper) even came out, and the original vote was set for Thursday, March 23rd.

Despite rolling back the mandate, cutting taxes for those making significantly more than even the San Francisco poverty line, Ryan lacked the support of the House Freedom Caucus (HFC), the eternal nemesis of the GOP leadership. The House Freedom Caucus is comprised of 25-33 GOP lawmakers from some of the most conservative districts in the country. They contended that the replacement bill did not go far enough in rolling back Obamacare, as it provided tax credits to help buy insurance, required insurers to comply with pernicious regulations like not being able to deny people with pre-existing conditions as well as continue providing essential health benefits (doctors visits, pregnancy, prescription drug coverage, etc.). The HFC prefers a three-pronged approach – repeal, replace, and improve. Any effort to replace the ACA must start with a repeal.

Speaker Ryan needed the HFC’s block of votes to pass the AHCA, and was forced to start providing concessions. To start, Ryan and Trump removed the requirement for insurers to provide a baseline of essential health benefits. Conservatives frequently use the talking point that a 65-year-old male should not have to pay for a plan that covers pregnancy or mammograms. Essential health benefits, they claim, keeps the minimum cost of insurance prohibitively high. Without these benefits, insurance companies would be free to create cheaper plans that would have fewer benefits. The thinking goes that cheaper plans draw in younger people to insurance pools, diversify risk, and lower the cost of premiums for the rest of the population. But there’s one problem — insurance companies have to cover those with pre-existing conditions, like addition. Aetna or Humana might find it more profitable to simply not offer any plans that cover addition at all. If they did, only people with addition would buy and that would create a very risky pool. This wasn’t a problem pre-Obamacare because Aetna or Humana could simply refuse those people. Put differently, if insurance companies are allowed to pick and choose which essential health benefits they have to provide (if any at all), they might choose to offer none. Only people that need a particular benefit would pay more for a plan that covers it.


Essential health benefits are quite popular, though — moderate Republicans were wary of their ill, pregnant, and aging constituents that might lose coverage or be forced to pay more under the AHCA. The more concessions Ryan and Trump provided to the House Freedom Caucus, the more moderate votes they lost. The HFC, emboldened, continued to extract further concessions but never got everything they were asking for. By Friday afternoon, Ryan had to pull the bill. After 7 years, he still did not have a piece of legislation that satisfied the various factions in the Republican party to pass the chamber he allegedly has control over.


The finger-pointing has already begun, with Trump blaming the Freedom Caucus, Paul Ryan, and Reince Priebus. While there is talk of another attempt, it is very unlikely that Republicans have the political willpower to stomach another healthcare related defeat. It’s more likely that they shift their focus to tax reform next. Lowering taxes on wealthy individuals and corporations is an argument that the House Freedom Caucus, moderate Republicans, the Cato Institute, the Koch brothers can all get behind, and Paul Ryan can all get behind.


Recommended reading







Promotional Codes

Uber – h3iv1 – $15 off first ride

Lyft – RISHI133 – $10 off first ride

I’m running an experiment where I try and purchase Google Ads against my Uber promo codes. So far, I’ve purchased about $60 towards my Uber promo code with roughly $120 – $140 in promotions redeemed. This is a good ROI, but Uber fares are low to begin with in San Francisco. In addition, each Uber redemption is worth $15 per ride. Most of my rides are between $5-$8. I’ll continue running these promotions until I can bring my costs down to less than $5 per ride. Last I checked, I was a little over $5.47 per ride.

How to Upsell Your Customers Using Data

This post originally appeared on the Umbel blog. The Umbel Marketing team helped edit and refine the original content.


As the New Year starts to take shape, we’ll be asked to reflect on the prior year and make resolutions for this one. In our personal lives as well as our professional lives, it’s important to think about what went well in 2015 and what we want to improve in 2016.

2015 was the year of analytics, marketing automation and first-party data. It was also the year of new media where brands and retailers tried new and innovative ways to engage their prospective customers. With all this focus (and money) on acquisition, it’s now time to figure out what to do with all those customers.

Some will inevitably churn as they came in through hefty promotions. Others will make a few purchases and stray elsewhere. But a select few will become loyal, passionate advocates of the brand. It’s time to plan for those and be a little bit more specific in the approach. It’s vital to engage these customers across channel, but it’s equally vital to make sure that they’re making incremental purchases and improving the bottom line.

Here’s a few ways to make it possible.

Continue Reading

The Fanatics, The Fairweathers, and The Friends

This post originally appeared on the Umbel blog. The Umbel Marketing team helped edit and refine the original content.


It’s a great time of the year to be a sports fan of any kind. Basketball season just started. Football season, collegiate and pro, is about halfway through. Baseball season is winding down. Oh, and Texas beat Oklahoma (full disclosure: I attended Texas).

With so much time and money spent by consumers on sports related activities, sports fans are surprisingly hard to reach. Perhaps it’s because (almost) everyone likes some kind of sport. Perhaps it’s because it’s an activity that marketers take for granted. Or, it’s because data in sports is notoriously hard to collect and parse.

At Umbel, we hear the latter a lot.

Continue Reading

How to Reach Maximum ROI with Customer Lifetime Value

This post originally appeared on the Umbel blog. The Umbel Marketing team helped edit and refine the original content.


Calculating customer lifetime value (LTV) down to the individual is not a new concept.

However, LTV is one of the most important metrics a business should track. So what is LTV exactly? LTV represents the amount that a particular customer is expected to spend over the duration of his or her relationship with a particular business.

For example, Wal-mart knows that I purchase online once every two months and spend $20 each time. Through a series of predictive models, Walmart can predict that my lifetime value is $89.25. The lifetime value metric incorporates how much it cost to acquire me, how much it costs to retain me, and discounts future cash flows.

Technology companies, including Umbel, are now calculating LTV in a more scalable way to help businesses establish a closer relationship to their customers and the channels where they were acquired. So now that Wal-mart knows that my LTV is an estimated $89.25, what strategies can they implement to increase this number?

Continue Reading

The Case for Product Management to Own the Roadmap

Most Product Management professionals believe that one of the responsibilities of the discipline is to own and control the product roadmap. In other words, Product Management decides (with the help of engineering, sales, client services, marketing, c-suite, and others) when to ship certain features. My assumptions were recently questioned when someone posed the question “Shouldn’t Product Marketing own the roadmap? They’re in charge of competitive analysis, differentiation, communication both internally and externally. Product Marketing should know what the product lacks, what competitors do better, and how the product needs to be better.” While conducting some preliminary research on my answer, I discovered that none of the blogs explain why Product Management owns the roadmap. They all take it for granted. Product Management should be in charge of the roadmap, and there are good reasons for it.

Continue Reading

Unify Brand, Agency, and Technology Efforts with Your Customer Data

This post originally appeared on the Umbel blog. The Umbel Marketing team helped edit and refine the original content.

Brands Were Thinking Big About Customer Data at the 2014 Digiday Brand Summit

We had a great time sponsoring the most recent Digiday Digital Brand Summit (#DigidayDBS) in Park City, Utah. Between hitting the slopes and accidentally blowing our per diem on a single meal (I’m looking at you Zoom), we attended a number of sessions and talked with some amazing marketers focused on how the most successful brands are reconnecting with their customers. Agencies and technology companies have done a great service over the past 15 years in helping brands make the transition to digital, but it’s clear that it is now time for brands to own and control their own consumer data. Leveraging customer data to unify brand, agency and technology efforts is an opportunity not to be missed by brand marketers in 2015.

As a brand, before you approach your agencies and technology vendors about planning the next campaign, I recommend doing some strategy work to see how you could be collecting and connecting customer data to improve marketing ROI. Stop to consider what your needs and potential opportunities are with respect to customer data. If your agency runs all of your campaigns, you could be missing out on some valuable data collection opportunities. Are there any initiatives that you’re taking in-house, like a large push for Valentines Day? Maybe that is where you can start. Has your agency (or agencies) been in charge of similar initiatives before? Your goal here is to create an itemized list to identify where data could make the biggest impact and improve your decision making while avoiding the political slowdown that sometimes encumbers a new initiative. Without such a list, the data deluge can become all too real, and you’ll find yourself going back to the traditional agency plan — even though you know deep down that you could be using data to increase conversions and build stronger 1:1 consumer relationships.

Continue Reading