Jeff Scott, Goldman Sachs Asset Management

Validating Ideas Through Strategic Partnerships

When Jeffrey Scott joined the treasury department of Microsoft in the early 1990s, he had worked for several years in a similar role at Dow Chemical, but he had never managed a multi-asset investment portfolio before.

So the first thing he did was look for help.

“I was young into my career when I took on the opportunity to lead Microsoft’s investment portfolio; I was in my early thirties,” Scott, who is now a Managing Director within the Global Portfolio Solutions Group of Goldman Sachs Asset Management, says in an interview with i3 Insights.

“Even though I had half a dozen years of Dow Chemical, managing derivatives and fixed income, it still did not prepare me to build a multi-asset portfolio.

“And when you are at that age, I think it’s important that you need to reach out to find mentors and to find help, but also you need some credibility to come with your ideas.

“Even though I had studied multi-asset funds, had a CFA [Chartered Financial Analyst designation], had experience, I didn’t have the experience of the folks that I reached out to.”

You can improve your governance structure dramatically if there is good visibility into what's happening in real time.

Scott’s efforts to reach out to specialists within the asset management firms morphed over time into what is now commonly known as strategic partnerships.

“What I would do is with my limited staff at the time was that we would do some modelling, come up with analysis that we thought made sense and then we ran it by our strategic partners. That was one concept,” he says.

“The other concept was that we’re moving into this multi-asset strategy space and we knew we were going to move up the risk spectrum in several asset classes and we had to try to figure out: ‘How do you do that? What is the right way?’”

Scott’s background in treasury departments taught him to focus on risk rather than return early on.

“Before Microsoft, I was in treasury at Dow Chemical and you’re managing risk, financial risk, that the company doesn’t want to bear – it could be interest rate risk, it could be currency risk, it could be commodities risk,” he says.

“So back in 1990 we were thinking about managing risk factors, or risk exposures, and to do that you had to be able to calculate what those risk exposures were. When I went to Microsoft in treasury and I had to start running a multi-asset portfolio, I said: ‘I have to somehow figure out what risk exposures I have.’

“I needed to see how asset managers think about investing in asset classes and making relative value decisions and I literally sat in strategy meetings with one of our strategic partners multiple times and they would let us listen in.

“Now, they didn’t do that for everybody, but again I was working for Microsoft.”

One of the issues Scott noticed early on was that asset managers had vastly more sophisticated risk monitoring systems than they had at Microsoft.

“Our strategic partners taught us how to invest money outside of front-end fixed income, to be quite frank, and I noticed that they were using risk systems that were more advanced than what we were using,” he says.

“We had siloed systems, like a yield book for bonds, but they had systems that were integrated with all asset classes.

“If I was going to move in that direction with the Microsoft portfolio, then I had to show the controllers, the chief financial officer and the board that we could understand and model the risk.”

I would behoove all investors who are managing public money or retirement money to have a more thoughtful diversified strategy that you base off a factor loading.

Not only would such a system help in managing risk, it would also be critical in developing a sophisticated governance framework.

“We needed a dashboard for two reasons: one is to manage risk and two, the board and the controllers needed to have a second pair of eyes to check what I’m doing: trust but verify,” Scott says.

“You can improve your governance structure dramatically if there is good visibility into what’s happening in real time.”

This emphasis on controlling risk quickly influenced the team’s thoughts about investing and they adopted a risk factor-based framework.

“What I had seen in the past is mean variance optimisation, but if you do enough homework you realise the flaws of mean variance optimisation and so I reached out to others to know how do you go about building that model,” Scott says.

“It really came down to diversification of risk factor exposures. And then there’s a bunch of math behind that to come up with the right weights.”

There are only a few pension funds around the world that have implemented a risk factor approach that spans the entire portfolio and one of the more cynical reasons given for this is peer risk.

After all, if markets move against you, you have less room to blame an external manager for the portfolio’s performance.

Scott acknowledges this problem, but says you can still implement most of the framework in the context of your peer group.

“To go completely diversified and have equal weight to, let’s call it, a dozen factors, then there will be a quarter, or a year, where you look foolish relative to your peers who have just one factor: equities, with maybe a little FX,” he says.

“So you have to calibrate your diversification relative to your peers, but you need to educate your board and your constituents about what you’re trying to accomplish.

“I started in 1990 with risk factors and, yes, it’s become more popular, but that is probably because it makes more sense.

“I would behoove all investors who are managing public money or retirement money to have a more thoughtful diversified strategy that you base off a factor loading.

“Not 350 factors, but eight to 10 that you can explain to your board and that you can build a governance framework around to say: ‘Okay, how far are we deviating relative to that factor?’”