In his book, The Halo Effect (Free Press, a division of Simon & Schuster, 2007), Rosenzweig discusses these errors in detail and what companies can do to avoid them. He recently spoke to Workforce Management New York bureau chief Jessica Marquez about the implications of the halo effect on HR and what managers can do to address it.
Workforce Management: What is the halo effect?
Phil Rosenzweig: The halo effect has to do with human beings’ general tendency to let an overall impression about something shape particular judgments. For example, if I think that an individual is a high-performing manager, I will probably look at various things they do and have a generally better impression of him/her. This happens because it is very hard for us to separately evaluate many different things about a person or a company. We tend to let these things flow together.
WM: You talk about how many management books apply the halo effect. What is going wrong here?
Rosenzweig: What they do is they gather data that are corrupted by the halo effect. They treat those data as if they are valid and objective data, but they are not. Imagine a company that is doing well. Revenues are up, profits are up and share prices are up. Most people will look at that company and say, "Wow, they have a brilliant strategy, their chief executive is a real visionary, they have great people, they listen to their customers," and so forth.
When that same company suffers a downturn, it’s very easy to make the opposite attributions. We say, "Oh, they lost their way, they became complacent, and they took the customer for granted," and so forth. Occasionally companies really do get worse in those kinds of ways, but usually they don’t.
Usually what’s happening here is that financial performance becomes a halo and it makes us make lots of other attributions and judgments.
The problem is when people try to do what they think is serious research about companies and the data that they gather is contaminated by the halo effect. Then they are using biased data and their conclusions will be based on that biased data. That’s the problem of so many big business best-sellers.
In a book like Jim Collins’ Good to Great, he congratulates himself on all of the data he has put together. But when you look at it, a lot of it is based on sources that commonly have the halo effect.
WM: What should HR executives take away from your book?
Rosenzweig: When it comes to human resource management, you want to ask things like "How many people who I make an offer to do I hire? How many midcareer people do I keep versus attrition?" You want to look at things that are objectively measurable, not simply attributions based on performance. Because if you simply ask, "Are my employees happy?" you can measure that. But whether satisfied employees drive performance or high performance drives employee satisfaction is very hard to untangle.
In my book, I mention a study that tries to untangle this. What they concluded was that high-performing companies lead to employee satisfaction more than employee satisfaction leads to high-performing companies. That’s actually not especially good news for HR managers.
HR managers probably say we want to keep our employees happy because that leads to high performance. And it probably does, but the opposite causality is probably even stronger. High-performing companies tend to have satisfied employees.
WM: Does that mean HR executives shouldn’t try to improve employee satisfaction?
Rosenzweig: No, they probably should. But they should understand that high performance also leads to employee satisfaction. Is it important to have satisfied employees? I think so, but it’s probably less the driver of performance than we commonly think it is.
WM: How then should HR executives persuade their CEOs to fund any of their programs if they can’t prove that causality?
Rosenzweig: If I was an HR executive, the initiatives I would focus on would not simply be "How do I have enthusiastic or satisfied employees?" It would be things like "How did I attract people of certain skills? How do I improve skill capability levels within the company and measure those things in an objective way?"
Then I would probably try to find ways to demonstrate through a longitudinal study that capability and skill building—and maybe to some extent, satisfaction—do have demonstrative effects on company performance. But you cannot do that by looking at simple correlation.
So one message to HR managers is, if you are trying to persuade people about the importance of what you are doing, that which you are doing should be more tangible and have more specificity than simply "Are our employees satisfied?"
WM: One metric HR executives use a lot is turnover. They say, "We launched this program to reduce employee turnover and in doing so we saved XY in costs, and thus, this initiative has contributed to business performance." Is that a fair line of causality to use?
Rosenzweig: It’s certainly better than a measure like satisfaction, because turnover is something that can be objectively measured. Either the people are there or they are not there. That’s a separate question if you are talking about the direction of causality. And there I think it’s true that high-performing companies probably have less turnover because, again, people like to be on a winning team. So do you improve performance by lowering turnover, or do you lower turnover by improving performance? It probably works in both directions.
WM: What does drive business performance?
Rosenzweig: Business performance is about two things: It’s about strategy and it’s about execution. Strategy is fundamentally about the choices we make during conditions of uncertainty. When you make choices during conditions of uncertainty, you are running a risk. Unfortunately, there is no way of being a high-performing company without making choices in conditions of uncertainty and running chances of failure.
A lot of people want the predictability and confidence of guaranteed results. But that’s just not the case in the world of business. A lot of these best-sellers that talk about how you can become great in a predictable manner clearly forget that companies need to do this.
The second part is the ability to mobilize resources, including human and other resources, to deliver on our chosen strategy.
Where does HR management fit in here? I think HR management is one of the most important aspects of execution. Business involves people. How do you attract people, build their capabilities so that they can make decisions and mobilize resources to most effectively execute the strategic choices that we have made? HR is absolutely central in that endeavor.
WM: Does this mean that HR will never get that seat at the table that they are always talking about?
Rosenzweig: No, they should have a seat at the table because there is no way that companies can be successful and execute well without terrific people who are able to do a great job. But if I was an HR manager, I would not try to get that seat at the table by saying, "Let’s have initiatives that make people happy, because that’s going to drive business performance."
I would say, "What are the skills that people need to execute well, to drive performance? And if we perform well as a company, guess what? People are going to be happier." The way you are going to get the leverage through HR is not through satisfaction, it’s through skilled people making good choices to drive high performance, and as a result, morale and attitudes will follow.
It’s a question of where HR is going to put its emphasis. I think if you emphasize things like business skills that are important to make the right choices in our industry, not only is that the right answer, but I think that’s going to gain a lot more respect and traction in the eyes of the non-HR managers who have their seat at the table.
WM: A lot of companies vie to get on lists such as Fortune’s Most Admired Companies or Best Companies to Work For. Should they not focus on that?
Rosenzweig: They probably should, because there is a wonderful halo effect if you do. And people will suddenly think that you are brilliant in a lot of ways. But the way you get on those lists is because to a great extent those perceptions are shaped by financial performance.
For example, last year ExxonMobil had unbelievably high profits. Why? They are a well-run company, but they benefited from a big spike in oil prices. But because they showed such extraordinarily high profits suddenly, people were saying, "Wow, they really have great managers, they are really innovative, they have great quality products."
I think if you look at ExxonMobil, they probably have the same managers as three years before and the products are not more innovative or higher quality. Why do we say all of these things? Because high performance creates a halo that causes us to make all of these inferences.