Authers’ Notes: Benjamin Ho on ‘Why Trust Matters’

Benjamin Ho: Transparency and humility are great ways to build trust. Game theory and lab experiments show that distrust breeds from lack of information and when promises go unmet. Transparency, authenticity, repeated positive interactions are the key building blocks I identify in my research.

That said, on the other hand (as an economist, there is always an “on the other hand”), building trust often comes at the expense of perceived competence. Too much transparency can reveal too much of the sausage; too much humility can signal lack of competence. I’ve found that sometimes attempts to build trust can lead to lower share prices and lost customers. Therefore it is always a balancing act.

So it’s an empirical question. Luckily for the Fed, we have a number that can help keep score. I think it is notable that interest rates have remained low through crisis after crisis, signals that, at the very least, markets expect the government will pay back its debts and that inflation will be transitory (although that has started to shift a bit).

John Authers: Obviously, the Fed is of great interest at present as President Joe Biden prepares to decide between Jerome Powell and Lael Brainard to lead it. I think that raises another question of trust. We need the Fed to be a trusted institution, both that it be trustworthy and that it be trusted. The Fed chair also needs to have substantial autonomy to act in a crisis. But how do they gain the legitimacy to do it? And does the process where the president chooses between two people who’ve been part of the Fed for years help give them the necessary legitimacy? You can devolve power to the Fed — but can you devolve legitimacy and trust?

Benjamin Ho: One of my favorite papers, “The Politician and the Judge” by economists [Eric] Maskin and [Jean] Tirole, discusses this dilemma that we need our government officials to be accountable (like the Politician) but we also need to give them discretion (like the Judge).

The Federal Reserve chair is a great example of the latter. We need to give them space to make the right choice even if it is unpopular, and hence the institution is designed with features like 10-year terms, and a history of apolitical appointments, where presidents regularly reappointed the chair from their predecessors.

That norm is currently being challenged. I actually have no problem per se with that development; I think there is good reason for more democratic accountability. But I do think it should be done with care and awareness of its consequences re: the trust in the Federal Reserve system.

On how best to preserve legitimacy of the Fed chair choice, I think we need to think of other markers of trustworthiness. The traditional markers have been fancy degrees, or perhaps bureaucratic reticence, but I think we are also moving to a time when we have questioned such traditional markers.

John Authers: Saying sorry: In your work you have highlighted that for apologies to be effective they need to be costly. In a digital world, the reputation of any consumer-facing entity can be scored on a continual basis. Is that ongoing monitoring and accountability a positive development? It would seem to offer early alerts for adjustment and avoid the “grand apology” outcome.

Benjamin Ho: Sometimes for trust to grow it needs room to breathe. Anyone who has seen a rom-com will know that the protagonists often have to learn to be vulnerable before they can get together. The contract theory literature finds that contracts between firms are often intentionally left underspecified to leave room for trust. In lab experiments, [Armin] Falk and [Michael] Kosfeld showed that in trust-game experiments, when parties exerted too much control and took too much risk away from the relationship, trust and cooperation declined.

Additionally, continuous reputation management can begin to feel artificial. As you note, apologies need to feel costly, and a mechanistic automated apology does not feel costly — it feels easy. And easy apologies are less effective. I do think an awareness of reputation is good, but I worry about overmanagement.

John Authers: Could I ask a little more about the role of money in apologies? Many years ago I covered the campaign on behalf of Holocaust survivors to retrieve money that had lain dormant in Swiss banks and other institutions. It was very important to them that the bankers apologize and take responsibility; it was also very important to get their money back. Many survivors felt conflicted between the two, and the banks worried that they could look as though they were trying to pay to make the problem go away. How important is money in making an apology meaningful?

Benjamin Ho: The role of money and apologies is a tricky one. It depends very much on the context. My main contribution to the literature is that the effectiveness of an apology is proportional to its cost.

In a recent experiment I conducted with Uber, where we tested different kinds of apologies on 1.5 million customers after they experienced a late ride, we found that a $5 coupon worked really well; customers appreciated the gesture and spent more than $5 more than those in the control group.

However, we found that an apology that was perceived as too cheap would backfire, leading customers to spend less than those in the control group. I think the same idea applies for more serious issues. The apology must be commensurate with the crime or it will backfire. How you weigh a dollar value against a life is perhaps an unanswerable question.

John Authers: One more thought on central bank accountability. Sir Paul Tucker, formerly deputy governor of the Bank of England, wrote a massive tome on the entire issue of how to devolve power while maintaining legitimacy. It’s fascinating that a senior central banker felt that this was such an issue. It’s called “Unelected Power: The Quest for Legitimacy in Central Banking and the Regulatory State.”

On crypto, a reader makes the following point: “The decentralized ledger technology behind certain cryptos reflects distrust in a system. I would counter that what is needed is a better trust-based system, not a distrust one. And somewhat ironically, many crypto advocates, who appear nihilistic about existing financial systems, really, really want to believe in a distrust-based system. That is notably paradoxical.” Do you agree, and how might we resolve the paradox?

Benjamin Ho: Agreed! I have my own biases, but I happen to think we’ve gotten good at building human institutions based on trust. The story in the book is about how we’ve built those institutions over thousands of years, and so far I don’t (at the moment anyway) see a need to move away from it.

The canonical blockchain application, Bitcoin, is based on a distrust of the financial system, yet markets still seem to trust the U.S. dollar and perceive U.S. Treasuries as the risk-free rate, and I trust the judgment of markets. The idea of wanting an alternative to traditional currencies untethered to political systems as a hedge anyway is reasonable, and is a typical justification for investing in gold, so I see Bitcoin as filling a similar niche. But I think the current financial system works reasonably well, and I’m not quite sure what problem Bitcoin is trying to solve (aside from bypassing laws and regulations).

I am sympathetic to partisans of blockchains in that human institutions are often besieged by bureaucracy and politics and potential corruption. And so I do see value in algorithmic ways to sidestep those things, although I also see traditional institutions pushing back and trying to re-regulate any transactions we move away from them. (Weber famously extolled the virtues of bureaucracy as the essential component of what makes institutions trustworthy, so human institutions are almost inevitably bureaucratic.)

Having consumed too much sci-fi (“Dune,” “Battlestar Galactica,” “Matrix,” “Terminator,” Asimov), my gut feeling is that bad things happen when we cede too much authority away from humans and toward algorithms. But I am open to seeing what the future brings.

Stacy-marie ishmael: A reader asks via Twitter: What is the connection between trust and the origin of money?

Benjamin Ho: I love this question! You can get dizzy trying to pin down just what money is, and a popular misconception about how money arose is that money was a way to improve the barter system. In fact, a better way to think about money is that money was a way to keep track of trust. In hunter-gatherer societies, a lot of economic life was governed by gift exchange. I shared my hunt as a gift, in the hope that you will share your hunt next time. As societies grew, it became more and more difficult to keep track of who owes who a favor. People began using markers to keep track of favors. The earliest markers became the earliest form of writing. Other markers developed into money.

Stacy-marie ishmael: Punishment: So many good papers to bookmark! I want to hear more about “giving folks space to make the right choice.” In your section on religion, you note that the notion of punishment is one thing that enables trust. This concept isn’t developed elsewhere. Could you say more about accountability as a mechanism that enables both trust and trustworthiness outside of the religious context?

Benjamin Ho: Great question! There is so much that could be said about punishment. A recurring theme throughout the book is that institutions build trust through two channels:

1. by creating incentives for good behavior, and
2. by helping to match people to trustworthy people.

Punishment does both. By lowering the payoff to bad behavior, it shifts the optimal strategies in trust games toward more cooperative behavior (i.e. it changes the incentives).

At the same time, living in a system that strongly incentivizes good behavior helps create more trustworthy people, as those good behaviors become internalized over time. We can see this in the macro literature where [Robert] Barro and [Rachel] McCleary show that beliefs in hell cause economic growth. A belief you will be punished internalizes good behavior, creating more growth and prosperity for everyone.

The problem with punishment, though, is, well, it is punitive. In a neoclassical model of punishment (e.g. Gary Becker’s classic “Crime and Punishment”), punishment is good for its deterrent effect, but punishment is bad because it is costly (more prosaically, you are taking a potentially productive worker out of the economy and paying for their imprisonment, but more fundamentally, you are depriving someone of their liberty).

Much of my career has been spent looking at apologies and the law, and one area where the two intersect is the movement toward restorative justice as an alternative to conventional punishment. Once a crime has been committed, the deterrent effect is moot, so punishment has no obvious benefit, especially because the empirical literature shows that time spent in jail may cause an increase in the probability of committing another crime. The restorative justice movement asks: Can opening a dialogue between victim and offender do better than punishment because it helps to restore trust?

The use of truth and reconciliation commissions after wars and conflicts is premised on the same idea.

John Authers: My impression is that the lack of punishment or “perp walks” after the global financial crisis had a critical role in undermining public trust in financial institutions, and the government, over the last decade. Without punishment being seen to be done, there was no sense of justice, and a strong sense that the bankers themselves had been bailed out — even though the political imperative had been to bail out their customers.

How might we apply your framework to the dilemma that faced the Barack Obama administration of how to go about prosecuting miscreants after the crisis?

Benjamin Ho: I personally am a big fan of forgiveness. Building and rebuilding trust is a two-way street. The transgressor needs to work on being more trustworthy, but the transgressed has a role in trying to empathize and forgive. We have a number of psychological biases that make forgiveness difficult; we have a hard time recognizing the situation we’re placed in and the systemic reasons for their decisions. I think it makes sense to work on that.

Of course punishments are also important in a just society, and for maintaining incentives for good behavior and the integrity of the system. So we need to balance all of those things on a case-by-case basis.

John Authers: Just to follow up on crypto: You wouldn’t see the growth of crypto itself as a sign of loss of trust in the Fed? I tend to agree that it’s a solution in search of a problem. But might the extent of interest in it show declining trust in the existing money (just as the surge in gold price back in 1980 showed a lack of trust)? And I suppose a broader question: Trust is evidently important, but how do we see it and measure it?

Benjamin Ho: I do think the genesis of Bitcoin and other cryptocurrencies was based on the distrust some have had for the Fed and the traditional money system. There have always been advocates for the gold standard, and anti-establishment types who have distrusted authority. I suspect that the growth of crypto has been driven primarily by speculators hoping to make money, and by entrepreneurs and innovators who do see promise in the potential technologies that can be built on new platforms.

As for how you measure trust, the classic way economists measure trust is via interest rates. We can back out mathematically, based on the interest rate that is charged, the probability that the parties involved think they will be repaid. For now, interest rates remain low, and I think that does tell us something.

Stacy-marie ishmael: Funnily enough, one of the “attractions” touted by various players in crypto who offer lending products is high interest rates!

Stereotyping: You made several useful references to the trade-offs of relying on stereotyping (also, Scott Page’s book “The Diversity Bonus” is so good). What do you think is the role of institutions and systems in helping allay this issue?

Benjamin Ho: We all hold stereotypes and biases. We can work hard to minimize them but they are always there. One of the founders of modern sociology, Max Weber, wrote around the beginning of the 20th century that bureaucracy is the key tool institutions have to eliminate favoritism. Rules and processes can overcome human fallibilities. At the same time, however, Gramsci, in the footsteps of Marx, argued that institutions are put in place by those in power, to enshrine and perpetuate their power. I think both are true.

So we need institutions, but um ... good ones, not bad ones? I started grad school with the dream of solving the problem of how to create good institutions, but have since concluded that we have a long, long way to go. I am very sympathetic, though, to the argument in James C. Scott’s “Seeing Like a State” that we should have humility in redesigning them.

Stacy-marie ishmael: Let’s dive deeper into the problems with stereotyping based on traditional markers of trustworthiness. As you note, that can lead to stereotyping. How can institutions help mitigate this?

Benjamin Ho: Traditional ways institutions use to avoid stereotyping are all valuable. Setting up systems that involve accountability and oversight, collecting data on diversity is crucial for transparency, a key component of building a trustworthy system, increasing awareness of potential biases are all valuable. And we see many of these systems in place.

I think one area where we can do more is to build institutions to bring together diverse people in new ways. I think part of the problem is we live in increasingly isolated bubbles. Our cities and online spaces and workplaces have all become in some ways more homogeneous.

I am very much heartened by research on trends in gay marriage (if you remember, not long ago Barack Obama and Hillary Clinton were both publicly against gay marriage), and the shift in public attitudes has been fast and dramatic. A key reason is that in recent decades, people became much more likely to encounter an openly gay person in their personal lives. And that helps us see people as people instead of an abstract group.

I think we need more of that — more institutions that help break down barriers. National service might be an extreme solution (which could be volunteer service like AmeriCorps or Peace Corps, not necessarily military service), but in places where you have mandatory national service, you do see more social cohesion.

John Authers: Peter Atwater of Financial Insyghts, who has taken part in book club blogs in the past, asks about the role of trust in broader markets: “Environments of trust are typically those where we have ceded control to others in exchange for increased certainty. I trust my plumber, for example, to fix my burst pipe. Today, many investors appear to have ceded control to the Fed — we hear ‘TINA’ and ‘don’t fight the Fed’ a lot. Are you concerned that we don’t have investor confidence today so much as investor compliance? Moreover, what will this mean for the Fed, if investors conclude their pipe still has a leak?”

Benjamin Ho: I love trusting your plumber being an example of how you place trust in someone for greater certainty in the outcome. When it comes to the Fed and the financial system, trust is paramount. The whole idea of money is built on trust.

I actually think our trust in institutions as a whole these days is strong. The distrust we see is mostly an increase in partisanship, which comes from greater ideological sorting. We don’t trust Congress but we trust our congressperson. We don’t trust “the media” but we avidly consume the media we trust. We don’t trust medicine, but we still trust our doctor.

Similarly, when it comes to the financial system, markets have been remarkably steady through many crises. It’s the partisanship I worry most about.

John Authers: I’d like to move on to the key question of trust and human and economic development. These fascinating charts from the book show that there is a correlation between levels of trust and economic growth, and also between trust and the rule of law.

Two key questions: First, how do you measure trust for this exercise? And second, why do some countries have higher levels of trust than others? In particular, China’s high level of trust is remarkable; how does Chinese society reach such a level of trust?

Benjamin Ho: Our measures of trust are imprecise. The simplest comes from the World Value Survey, where they poll people on whether they agree with the statement that you can generally trust the people around you.

But economists prefer a measure that has more tangible payoffs (like TV’s Dr. House, experimental economists assume people lie unless real money is on the line), where we offer people some amount of money, say, $10, and give them a chance to “invest” it in an anonymous person from the same village for the chance of increasing their investment to $15, but at the risk of losing it all if the other person proves to be untrustworthy. These experiments have been conducted around the world and help construct these indices.

Where this trust comes from is complicated. Culture and religion play an important part. Past work to show that trust causes economic growth uses religion as a natural experiment. The idea is that what religion a country has is based on historical factors (often invasion) that are unrelated to other determinants of growth. They find evidence that these historical influences are a major determinant of trust in countries today.

John Authers: On religion and trust: Does this tend to mean that Max Weber was right to link the rise of successful capitalism to the Protestant ethic?

Also, digging back a long way, the British left-wing thinker R.H. Tawney wrote a classic book called “Religion and the Rise of Capitalism,” showing how the two were inextricably linked. If religion builds trust, and trust is essential for capitalism and markets to work, does modern experimental psychology research suggest that Tawney was right?

Benjamin Ho: I think there is truth in that. The empirical economic literature has actually come a long way since the days of Weber and Tawney. [Daron] Acemoglu and [James] Robinson have an excellent book on the historical precursors of successful countries today. Their seminal research links the presence of germs that affected European settlers in the 1500s to economic growth today, based on the idea that the institutions European settlers brought with them when they settled have long-lasting impacts.

Related work has looked at the historical impact of Protestant versus Catholic colonization. Going back even further in history, anthropologists have looked at how the invention of moralizing gods led to the development of pre-modern civilizations.

Of course, placing too much confidence in any empirical research over so many centuries is fraught at best. But given how fundamental trust is in markets — we see it all around us — I think there is reason to believe it is.

Stacy-marie ishmael: Thanks so much for this fascinating discussion! My copy of the book is all underlines and highlighting, so I wanted to ask you something that, for me, is at the heart of what makes this so interesting: What do you think is the biggest difference between the economic understanding of trust that you’ve been describing and the societal understanding of trust that we’re experiencing?

Benjamin Ho: One big caveat for the book is that I am very much an economist, and the book is very much an economic one (even though I have been very much influenced by work in sociology, anthropology, philosophy, etc.).

One thing economists miss is that we try very hard to isolate specific factors, to find the signal within all the complexity and noise. But often the complexity is the most important part of the story.

The other thing economists miss is that we tend to presume that people do things for a reason. They weigh the costs and benefits of their actions (“Do I trust somebody or not, do I behave in a trustworthy manner or not?”) and make a choice based on the trade-offs.

I think other social sciences tend to see things differently. A caricature of this I got from a professor in grad school is that a psychologist might be more concerned about the influences of your past — how did events of your childhood influence your actions today? A sociologist is interested in the role of your place in society today, the social network and culture you are embedded in. Economists tend to focus on the future, assuming we base our choices only on their future consequences, calling everything else a sunk cost.

Of course that is an oversimplification, but a fair critique, and so I think a full understanding of the trust requires understanding our best (e.g. the anthropological record) and also the culture we are embedded in.

John Authers: Thank you for a fascinating discussion. Now it merely remains to introduce the next book we’re going to discuss, which is “Shutdown: How Covid Shook the World’s Economy” by Adam Tooze.

You’ll be relieved to know that this isn’t yet another rehash of the arguments about masks, vaccines and social distancing. Rather, it looks at the unambiguous fact that the world economy had an unprecedented peacetime shutdown last year, and examines how that happened, how different actors in the economy tried to deal with it, and what the consequences will be.

Time will no doubt change a lot of our judgments on this, but it’s a great and courageous attempt to write the first draft of history, on a subject we certainly need to be discussing, from a great professor at Columbia University.

We’ll aim to have one last book club live blog for the year next month, before Christmas. Watch this space, and trust us on this one. Thanks to everyone for taking part.

Bloomberg News provided this article. For more articles like this please visit bloomberg.com.

Read more articles by John Authers