Marina Gorbis's Blog, page 1573

What to Ask Your "Numbers People"

If you're a manager working with the analysts in your organization to make more data-driven business decisions, asking good questions should be one of your top priorities. Many managers fear that asking questions will make them appear unintelligent about quantitative matters. However, if you ask the right kinds of questions, you can both appear knowledgeable and advance the likelihood of a good decision outcome.

In my new book (co-authored with Jinho Kim) Keeping Up with the Quants, and in a related article in this month's HBR, we list a lot of possible questions for various stages of analysis. But in this short article, I thought it might be useful not only to mention a couple of the most important questions you can ask about data, but what some of the ensuing dialogue might involve.

1.Questions about Assumptions

You ask: What are the assumptions behind the model you built?

You think in response to their answer: If they say there are no particular assumptions, you should worry — because every model has assumptions behind it. It may be only that you're assuming that the sample represents a population, or that the data gathered at a previous time are still representative of the current time.

Follow-up: Is there any reason to believe that those assumptions are no longer valid?

You think in response: You are really looking only for a thoughtful response here. The only way to know for sure about whether assumptions still hold is to do a different analysis on newly-gathered data — which could be very expensive. Perhaps a particular relationship only holds when the values of a variable are moving in a particular direction (e.g., "this mortgage risk model only holds true when housing prices are going up — nah, that could never change!").

2. Questions about Data Distribution

You ask: How are the data you gathered distributed?

You think in response: If the person can't describe the distribution, he or she is a shoddy analyst. Good analysts should have already looked at — and be able to show you — a visual display of the distribution of your data on any particular variable.



If you are interested in one variable as a likely predictor of another, ask for a "scatterplot" and look to see if the data line up in any linear pattern; that would indicate a strong correlation between the two variables.

Follow-up: Do the data follow a normal distribution?

You think in response: If the analyst says that the data aren't distributed normally (i.e., in a bell-shaped curve), then he or she needs to employ different types of statistics (called "nonparametric" statistics), and some commonly-used ones like standard deviations and correlations don't apply.

You might ask how they adjusted their analysis based on the distribution. For example, nonparametric tests often require a larger number of cases for the same level of statistical confidence.

Second follow-up: Were there any significant outliers?

You think in response: If the data are normally distributed but there are some outliers (unexpected values that don't fit the pattern), you could ask what they might mean, and what the analyst plans to do with them. In some cases it may be reasonable to delete outliers — if, for example, they are the result of coding errors.

You get the picture. It's important to show with this dialogue that you are interested, somewhat knowledgeable, and dedicated to a good decision outcome. You're not trying to suggest with such questions that you know more than the analyst, or that the analyst is hiding anything from you. It's the same sort of conversation that a CEO might have with a division manager who is presenting financial results. Gentle probing is the desirable tone.

What Would You Do Differently If You Could See Yourself 20 Years Older?

Undergraduates who had gazed at their 40-year-old selves in virtual "mirrors" were 74% less likely to cheat for extra cash on a subsequent trivia test, says a team led by Jean-Louis van Gelder of the Netherlands Institute for the Study of Crime and Law Enforcement and Hal E. Hershfield of NYU. This and another experiment suggest that one reason people make self-defeating choices such as engaging in unethical behavior is that their ability to imagine their future selves is limited. They're less inclined to indulge in illegal acts if they can see vivid images of themselves such as the computer simulations presented by the researchers. See Hershfield's "Defend Your Research" interview in the June 2013 HBR.

The Brighter Side of Decades of Disappointing Investment Returns

The prospects for investors, be they corporate pensions, sovereign wealth funds, or individuals, appear bleak. John Authers, in Thursday's FT, musters up the evidence (or at least the opinions) that bonds and stocks are both likely to deliver paltry returns for years to come. Sheelah Kolhatkar, in a cover story (and what a classy cover it is) in the new Bloomberg Businessweek, argues that the great alternative to plain-vanilla equity and debt investing — the hedge fund — is more or less over, too. And London Business School's Elroy Dimson, Paul Marsh, and Mike Staunton, in a Credit Suisse report cited by Authers that's a few months old but it is well worth downloading and taking the time to read, make a pretty convincing case that the equity and bond returns that came to be perceived in the U.S. as "normal" in the decades following World War II — and particularly since 1980 — are anything but.

We're all pretty well aware of the negative consequences of low investment returns, although perhaps not of the specifics. Dimson, Marsh, and Staunton calculate that a 25-year-old entering a defined-contribution retirement savings plan — such as a 401(k) in the U.S. — needs to set aside 16% to 20% of her income (!) if she hopes to retire at age 65 at half salary.

Still, where there are losers, there are winners. "While a low-return world imposes stresses on investors and savers in an over-leveraged world recovering from a deep financial crisis," Dimson, Marsh, and Staunton write, "it provides essential relief for borrowers." The biggest borrowers these days are governments, and because governments possess powers that most borrowers don't, they are often able to cut their borrowing costs at the expense of creditors through a variety of techniques that Harvard economist Carmen Reinhart dubs "financial repression."

Financial repression doesn't sound like a good thing, and in some ways it isn't. But the great global recovery after World War II was accomplished in an environment of artificially low interest rates and capital controls — that is, financial repression. So it doesn't have to be bad economic news. Sometimes governments actually use the money they borrow wisely.

Then there are the other tradeoffs inherent in investment returns. Over time, Rob Arnott and Peter Bernstein demonstrated in a great 2002 article in the Financial Analysts Journal, stock market returns should and do trail economic growth. That's because:

Shareholders can expect to participate only in the growth of the enterprises they are investing in. An important engine for economic growth is the creation of new enterprises. The investor in today's enterprises does not own tomorrow's new enterprises — not without making a separate investment in those new enterprises with new investment capital.

So when U.S. stock market returns outpace economic growth, that money has to come from somewhere. It can be taken out of workers' paychecks, creditors' pockets, or government coffers. It can come from overseas, as U.S.-based companies take advantage of faster growth elsewhere. And it can come from the future, as investors bet on faster growth to come.

During the great bull market of 1981 to 2001, all of these factors played a role. I'll focus on workers' paychecks. Labor's share of income actually didn't drop all that much over those two decades — in part because of big increases in worker pay (including a lot of stock-option jackpots) in the final years of the 1990s boom. But the longer-term trend since the 1960s has been of a significant decline in the percentage of total income going to workers, one that has accelerated over the past decade — during which stocks have sputtered but bond investors have done quite well.

So here's the big if. If those low investment returns that everybody's projecting are low just because the global economy keeps sputtering, that's not good news for anybody. If, however, they signal a regime change in which the financial sector's great rise over the past few decades begins to reverse and other sectors — governments and workers mainly, although there are certainly business sectors that have lagged as well — grab a greater share of economic growth, that doesn't have to be a bad thing at all.

To a certain extent, I'll admit, this amounts to saying that if times are bad they'll be bad, and if they're good they'll be good. But it's important to draw the link between two important and seldom-connected discussions: one about the prospects for investment returns, the other about the proper role of finance and capital in the economy. Because if the financial sector shrinks in relative size, and working stiffs gain a greater share of national income — both of which would be healthy developments, I think — the share of economic growth going to investors will have to decline.

The Booming Business of Craft Cocktails

An interview with Thomas Mooney, co-owner and CEO of House Spirits Distillery.

Download this podcast

A written transcript will be available by July 18.

For U.S. Employers, Health Care Reform Is a Watching and Waiting Game

Understanding the implementation plans for the health care law passed in 2010 is a complicated job for even the most seasoned policy wonks. A cascading series of deadlines, operating systems, and reporting rules have to be arranged, tested, and communicated. Add to this the resounding refrain that many Americans — and perhaps even many employers — don't fully understand what compliance with the landmark law will mean.

These logistics took a sharp new turn last week when the Obama administration decided to delay enforcing mandatory employer and insurer employment requirements until 2015. This provision would have required most companies with the equivalent of 50 or more full-time workers to offer health benefits on January 1, 2014, or pay a fine of at least $2,000 per employee. (Click here for a chart that breaks out the employer responsibilities around the Affordable Care Act). The delay has spurred fresh debates about the law and its oversight. It has also led GOP leaders to start advocating anew for rollbacks or delays on other key provisions, like the individual mandate. The House could call as many as three votes on the ACA in July.

The Treasury Department blog post announcing the change cast the move as careful implementation of the law that recognized "concerns about the complexity of the requirements and the need for more time to implement them effectively."

There's a key word: time.

Most large U.S. companies already provide workers with health benefits. But some of the businesses that would have been impacted by the employer mandate — in sectors like and food service — didn't appear to be ready to comply with the law. The administration also needs time to work on the reporting requirements that will be part and parcel of making the system work.

So the law has hit a delay, political heat is rising, businesses are in the middle, and many people are still just plain confused about what they have to do.

The tumult puts the broader role of the employer as conduit to health care benefits into new context — and by most accounts in the midst of a sea change. Companies are reexamining their place in the health care chain and what that means both for their workers and their tax strategies. One trend worth noting is toward consumer-directed health plans, which put workers in control of spending and coverage decisions.

"If fifteen years ago we were to have a discussion of health care in one of our membership meetings, 100% of the people attending the meeting would say it's our duty and responsibility to provide health care to our employees," says Jeffrey McGuiness, CEO of the Washington-based HR Policy Association, which represents chief human resource officers from large companies. "Now, my sense is that a very large percentage of those same companies are saying, is that case any longer? Can we really do that in this environment?"

What happens next? Here are a few things to watch:

HR departments. Companies of all sizes are evaluating their health care strategies and objectives now. HR executives are finding health care planning to be as high on their list of priorities as talent management. But while businesses need to start strategizing, policy experts like McGuiness caution against making any long-term commitments until the road ahead for the law is clearer.

The definitions of full-time and part-time workers. The health care law defines a full-time employee as one who works an average of 30 hours a week. There are bipartisan bills afoot in Congress that seek to redefine a full-time worker as someone who works 40 hours, not 30. It's tough to say if these will go anywhere. But expect the issue of who is and isn't a full-time worker to keep coming up in the conversation.

What happens on October 1. That's when federal and state-run healthcare exchanges, where individuals and small businesses can shop for coverage, are due to be open for enrollment. It's a critical day for the government on the path to making the law a reality. If this deadline isn't met, the January, 2014, deadline for the individual mandate could get thrown into question. Uninsured young people are expected to be among the first to get the pitch to sign up (although they apparently won't be spiking any footballs to celebrate).

Growth in the U.S. health care sector. You can slice the news about growth in the health care sector many different ways. First, there is the eye-popping growth in jobs in the health care field, which defies most other charts you've seen lately on labor. Second, there are the long-running concerns about the high costs of health care in the U.S., something reform advocates have long hoped to see better calibrated. If we are on the verge of changes in the health care sector, these indicators may help track them.

There are many resources available to help keep up with or understand the law, some with particular value to HR leaders, or reporters, or consumers. If you have a useful one to recommend for managers, please leave it in the comments.

And if you'd just prefer a short refresher on what the law might look like for you in video form, try this one from the Wall Street Journal:

This post was updated at 5:30pm ET on July 11.

Are You Data Driven? Take a Hard Look in the Mirror.

The term "data driven" is penetrating the lexicon ever more deeply these days. Data Driven was the title of my latest book, and recent academic work shows that companies that regard themselves as "data driven," are measurably more profitable than those that aren't. So becoming data-driven is clearly a worthwhile endeavor. Yet for all the attention, I've yet to see any clear criteria by which leaders can benchmark themselves and their organizations to figure out what they need to do better.

In my view, the essence of "data-driven" is making better decisions up and down the organization chart. Over the years I've had the good fortune to work with plenty of individual decision-makers and groups, some terrific and some simply awful. From that work, I've distilled twelve "traits of the data-driven." (It bears mention that the data-driven also avoid some self-destructive traits; I'll take these up in another post.)

Traits of the Data-Driven

The data-driven:

Make decisions at the lowest possible level
Bring as much diverse data to any situation as they possibly can.
Use data to develop a deeper understanding of their worlds.
Develop an appreciation for variation
Deal reasonably well with uncertainty
Integrate their ability to understand data and its implications and their intuitions.
Recognize the importance of high-quality data and invest to improve.
Are good experimenters and researchers.
Recognize that decision criteria can vary with circumstances.
Recognize that making a decision is only step one.
Work hard to learn new skills and bring new data and new data technologies (big data, predictive analytics, metadata management, etc) into their organizations.
Learn from their mistakes.

All of these traits are important. And most are self-evident. Only a few require further explanation. First, data-driven companies work to drive decision-making to the lowest possible level. One executive I spoke to described how he thought about it this way: "My goal is to make six decisions a year. Of course that means I have to pick the six most important things to decide on and that I make sure those who report to me have the data, and the confidence, they need to make the others." Pushing decision-making down frees up senior time for the most important decisions. And, just as importantly, lower-level people spend more time and take greater care when a decision falls to them. It builds the right kinds of organizational capability and, quite frankly, appears to create a work environment that is more fun.

Second, the data-driven have an innate sense that variation dominates. Even the simplest process, human response, or most-controlled situation varies. While they may not use control charts, they know that they have to understand that variation if they are going to understand what is going on. One middle manager expressed it to me this way, "When I took my first management job, I agonized over results every week. Some weeks we were up slightly, others down. I tried to take credit for small upturns and agonized over downturns. My boss kept telling me to stop — I was almost certainly making matters worse. It took a long time for me to learn that things bounce around. But finally I did."

Third, the data-driven place high demands on their data and data sources. They know that their decisions are no better than the data on which they are based, so they invest in quality data and cultivate data sources they can trust. As a result, when a time-sensitive issue comes up they are prepared. High-quality data makes it easier to understand variation and reduces uncertainty. Success is measured in execution, and high-quality data makes it easier for others to follow the decision-makers logic and align to the decision.



Further, as one executes, one acquires more data. So the data-driven are constantly re-evaluating, refining their decisions along the way. They are quicker than others to pull to plug as when the evidence suggests that a decision is wrong. To be clear, it doesn't appear that the data-driven "turn on a dime"; they know that is not sustainable. Rather, they learn as they go.

Now take that hard look in the mirror. Look at the list above and give yourself a point for every trait you follow regularly and half a point for those you follow most — but not all — of the time. Be hard on yourself. If you can only cite an instance or two, don't give yourself any credit.

Unless you're one of the rare few that truly score seven or more, you need to improve. While each person and organization is different, I'd first recommend starting by pushing decision-making down the organization. I've already noted the benefits. It may be tough and counterintuitive, especially for managers that want to feel in control, but it's worth the effort.

Second, invest in quality data. Frankly, you simply cannot be data-driven (or do anything consistently well for that matter) without high-level of trust in your data and data sources. You're reduced to your intuition alone, the antithesis of the goal here. Quality data is a necessity.

Now, take one more step. You've taken a hard look at yourself. Engage your management team in doing exactly the same thing for your organization.

How Share-Price Fixation Killed Enron

In December, 2001, just prior to filing for bankruptcy, Enron Corporation had approximately $2 billion in cash and no debt coming due. Despite its infamous financial chicanery, it still appeared to be a viable, profitable firm. So why did Enron go bankrupt? Was it because of the fraud, or was there another reason?

At the annual conference of the Association of Certified Fraud Examiners late last month, former Enron Chief Financial Officer Andrew Fastow, who served six years in prison for his part in Enron's deceptions, offered an explanation. In a keynote speech, he said Enron went bankrupt because of "decisions" made in October 2001. He didn't say which decisions. But after hearing Fastow speak twice to my Financial Statement Accounting class and reviewing independent evidence, I think I have good idea. It appears that Enron's final fatal mistake was to try to support its stock price instead of living up to key contractual obligations required to maintain its credit rating.

Enron owned the largest natural gas pipeline system in the U.S., was the largest trader of natural gas and electricity, owned the largest wind power company, and owned a large electric utility in the Northwest. These divisions all generated consistent earnings and cash flows. Enron also owned two "prospective" businesses: Enron Broadband (the first company to offer live video streaming and one that was establishing the largest "cloud storage" system at the time) and Energy Services (providing services to help other companies make their facilities more energy efficient) that weren't producing earnings but weren't a significant drain on the company in late 2001 and, at least with hindsight, represented significant opportunities. The company also had three divisions — Water, International, and Merchant Investment — that were saddled with underperforming and over-valued assets.

What caused Enron to go bankrupt?

What caused Enron's bankruptcy was, quite simply, the loss of its investment-grade credit rating. Without investment-grade status, counter-parties in its trading business (its largest and most profitable segment) either refused to trade with Enron or demanded collateral (which Enron could not post). The loss of the investment-grade designation also accelerated other debt maturities.

So, what caused Enron to lose its investment grade rating? Were the problems at International, Water, and Merchant Investment too big to overcome? No. Were the rating agencies aware of Enron's oft-maligned financing structures? Yes. Did the rating agencies understand that the acceleration of debt maturities brought on by a downgrade could cause a bankruptcy? Yes.

Enron was rated BBB+ (or the equivalent) by all three rating agencies, which typically include all off balance sheet debt when determining a rating. Enron had created multiple non-consolidated Special Purpose Entities (SPEs) that were levered 97/3, meaning $97 of debt to each $3 of equity. Enron's court-appointed bankruptcy examiner estimated the SPEs comprised $14 billion of off-balance sheet debt. Adding the SPEs to Enron's balance sheet would cause Enron to lose its investment-grade rating.

Enron's solution was to alter the nature of its SPEs. A typical SPE requires a company to make cash payments to the SPE if its assets fall in value. Enron created Contingent Equity Vehicles (CEVs) wherein Enron pledged to issue new equity, rather than cash, in the event of asset impairment. By way of illustration: including the CEVs on the balance sheet adds $14 billion to assets and $14 billion to liabilities. In the worst-case scenario of a 100% impairment of the CEVs' assets, Enron's assets and equity (retained earnings) would then fall by $14 billion. However, Enron then could essentially convert the $14 billion CEV debt into equity by issuing new shares. The net result is a drop in assets and debt (equity falls with the decline in assets but goes back up with the issue of new equity) to Enron's exact balance sheet position without the CEVs. This is why the rating agencies could exclude the SPE debt.

The key feature of these CEVs is that they required Enron to issue the new equity, and they required the lenders and other counterparties to accept new equity in lieu of cash. Firms are often unable to issue new equity at just the moment they need it most, but here Enron could. In essence, these financing structures were a "fail safe" designed to ensure that Enron's balance sheet remained investment grade.

Enron's unpleasant choice

As the value of the assets in the SPEs became impaired, Enron faced an unpleasant choice: issue new equity as promised, thereby diluting current shareholders and causing a drop in stock price, or risk a loss of Enron's investment-grade rating and potentially destroy the firm.

Enron's CEO at the time, Ken Lay, decided against issuing the stock (and against living up to the financial structure created by Fastow), apparently believing that (a) dilution would cause an even greater loss of confidence than would the impairment of Enron's balance sheet from including the SPEs and (b) the rating agencies would back down. Instead, the rating agencies downgraded Enron, the trading operations were forced out of business, $4 billion of debt was accelerated, and Enron was forced to file for bankruptcy.

The stability of share price is a metric many managers and investors look at when evaluating the "quality" of a firm. However, in situations where a firm must maintain access to capital markets (e.g. a rapidly growing firm) or must maintain an investment-grade rating for contractual or business purposes (e.g. a financial firm), trying to manage a firm's stock price is clearly secondary to maintaining its credit rating (i.e. it is imperative to maintain the trust of the rating agencies). The financing structures built to protect Enron in just such an event were unwound. Ironically, the company's bankruptcy might have been avoided had Enron lived up to the promises in those oft-maligned financing structures.

Avoid the Deadly Temptations that Derail Innovators

Any promising new initiative — a stand-alone business venture or an innovation in an established organization — hits roadblocks and unexpected obstacles. Recently I've advised entrepreneurs and innovators about a different, seemingly better, dilemma: pop-up opportunities that look like short cuts to success. Too often, these turn out to be deadly temptations.

Consider these cases (with names disguised to protect confidentiality):

Bill's venture capital-backed business concept was to operate a new revenue-producing service for large U.S. professional organizations. In its first year, the venture landed two almost-committed pilot sites and a prospect pipeline for a multi-billion-dollar market. But almost at the same time, Bill was offered a lucrative deal to build a similar service for an English-speaking country outside the U.S. Feeling that the money was good and the chance to show credibility to U.S. customers even better, Bill took the deal, brushing aside numerous challenging differences and departures from his model. Then he was offered an even bigger international site in a developing country eager for American know-how, in partnership with a U.S. organization that could also be a customer. His financial backers urged him to take it — it would mean more revenue, fast. Suddenly Bill was in a different, less appealing business, jeopardizing building the U.S. business.

In this story, that giant sucking sound you hear is the draining of time and energy from the core business by tempting almost-related opportunities. The danger comes from possibilities that are close enough to be plausible but take attention away from the building the main business and don't prove the concept anyway.

Mary's temptations were slightly different but had similar consequences. She started a non-profit organization with lavish foundation funding and a high-profile board in order to spread an innovation in health care. This was a situation devoutly to be wished for by most social causes, but it proved limiting and almost fatal to Mary's project. The staff proliferated without clarity of purpose, and the organization became insular, looking inward and feeling they must be at the top of their field. Other groups courted the organization because it could bring funding, not because of a commitment to the innovation. Soon the goal became how to raise money, not how to support and improve the innovation. The organization took some government contracts to provide a somewhat-related but more routine service. Donors became confused about what the organization did. Private funding declined. The venture was in peril.

Ironically, these problems come along with looking like you might succeed. Investors, donors, potential partners, or bosses shower temptations on entrepreneurs who show promise. Joe's first wildly successful conference, accompanied by highly creative marketing, drew offers to him from investors who wanted to back him to go national, people who wanted to hire him to popularize their work, and companies that wanted him to be a distributor for merchandise sales. Joe was dazzled by the big-name people interested in him. But none of this built Joe's brand. The graveyard of startups is filled with innovators lured by glamour to let others take over their concept before it was fully developed.

In the western classic The Odyssey, Odysseus put wax in his sailors' eyes and tied himself to his boat to avoid being tempted by the sirens and lured into lethal rocks. In her new book Sidetracked, my HBS colleague Francesa Gino outlines ways to handle more contemporary distractions. Entrepreneurs who want to avoid the deadly temptations I identify here can take these actions:

Establish principles by which opportunities will be judged. Creating new initiatives benefits from the flexibility to improvise, as I've written, but boundaries and direction ensure that efforts add up in a coherent way and can be replicated and scaled. Strategy is what you don't do, not just what you do, as my HBS colleague Michael Porter has said.

Prove the concept you want to prove. Most people are concrete thinkers who will assume that a project is whatever they first see — why Bill's strategy for a prototype was very risky. It's important to build into the first model at least one glimmer of everything you anticipate for the full product, while screening out anything that doesn't signal future aspirations. For example, if you want corporate partners eventually, get at least one before you start. If you want to reach full potential in the domestic market, hold off on international forays. Sometimes walking away from money is smart strategy if it comes with unrelated requirements.

Control your identity. Put the right words around the project, and stick with them. Observers often reduce innovations to familiar elements, using language they already have, but which might not fit the initiative, leading to offers of distracting opportunities when the core business isn't understood. One innovation group developed a glossary of terms to be used, and words to be avoided. The same group also declined an investment from a source that would have sent misleading signals about the business the venture was in.

Don't lean insular. Innovators can lean in so far that they become insular. talking only to those that agree with them or flatter them. Was Kodak's demise precipitated by being Rochester-centric, where they were top of the heap, rather than mingling more in Silicon Valley where people had different views of the future of photography? Bill, Mary, and Joe were so flattered by money that they didn't check with outside experts who would have warned them of the dangers ahead.

In short, to get to where you want to go, ignore the deadly temptations that might spring up on an innovation journey. Stay focused on the purpose and the destination.

Smaller, More-Homogeneous Research Groups Are More Productive per Researcher

Research organizations like to increase team size to bring in new ideas, but a study of 549 research groups shows that in teams consisting of people from multiple disciplines, the published output per researcher in 13-person groups was 42% lower, on average, than in five-person groups, says a team led by Jonathon N. Cummings of Duke University. The greater the heterogeneity in disciplines, the less effective it was for groups to increase their size. In interviews, team members in large, heterogeneous groups complained of lack of familiarity and personal chemistry with colleagues and problems in communication, the researchers say.

Alarming Research Shows the Sorry State of US Higher Ed

It's dismaying how easy it is to screw up college.

I don't know exactly when, why, or how it happened, but important things are breaking down in the US higher education system. Whether or not this system is in danger of collapsing it feels like it's losing its way, and failing in its mission of developing the citizens and workers we need in the 21st century.

This mission clearly includes getting students to graduate, yet only a bit more than half of all US students enrolled in four-year colleges and universities complete their degrees within six years, and only 29% who start two year degrees finish them within three years. America is last in graduation rate among 18 countries assessed in 2010 by the OECD. Things used to be better; in the late 1960s, nearly half of all college students got done in four years.

Have graduates learned a lot? In too many cases, apparently not. One of the strongest bodies of evidence I've come across showing that students aren't acquiring many academic skills is work done by sociologists Richard Arum and Josipa Roksa and summarized in their book Academically Adrift: Limited Learning on College Campuses and subsequent research.



Arum, Roksa, and their colleagues tracked more than 2300 students enrolled full time in four-year degree programs at a range of American colleges and universities. Their findings are alarming: 45% of students demonstrate no significant improvement on a written test of critical thinking called the Collegiate Learning Assessment (CLA) after two years of college, and 36% improved not at all after four years. And the average improvement on the test after four years was quite small.

Consider a student who scored at the 50% percentile as a freshman. If he experienced average improvement over four years of college, then went back and took the test again with another group of incoming freshmen, he would score only in the 68th percentile. The CLA is so new that we don't know if these gains were bigger in the past, but previous research using other tests indicates that they were, and that only a few decades ago the average college student learned a great deal between freshman and senior years.

These declines in learning and graduation rates come during a time of exploding costs. the Pew Research Center found that the price of a private college education tripled between 1980 and 2010, and that average student loan debt for bachelor's degree holders who had to borrow was more than $23,000 in 2011. This debt is not dischargeable even in bankruptcy, and is certainly not erased if you fail to graduate.

Smart students from affluent homes and elite colleges and universities continue to do really well, but the rest of higher ed is sliding backward. Why is this? As was the case with the sub-prime crisis and subsequent economic meltdown, there is plenty of blame to go around. Many non-elite colleges have seen their enrollments jump in recent decades without similar increases in budgets, so resources per student have declined.

It also seems, though, that colleges in general have stopped asking students to work as hard, and the students have been more than happy to take them up on that offer. Arum, Roksa, and their colleagues document that college students today spend only 9% of their time studying (compared to 51% on "socializing, recreating, and other"), much less than in previous decades, and that only 42% reported having taken a class the previous semester that required them to read at least 40 pages a week and write at least 20 pages total. They write that "The portrayal of higher education emerging from [this research] is one of an institution focused more on social than academic experiences. Students spend very little time studying, and professors rarely demand much from them in terms of reading and writing."

Here's my advice to recent high school grads (and their families): don't be part of this shameful and lazy bargain. Resolve to work hard, take tough classes, and graduate on time. Many changes are necessary in higher ed, most of which will take a great deal of time. But the most effective interventions can start the day you show up on campus. Crack the books, find good teachers, and take the education part of your education seriously.

Arum and Roksa found that at every college studied some students show great improvement on the CLA. In general, these are students who spent more time studying (especially studying alone), took courses with more required reading and writing, and had more demanding faculty. So the blueprint is here. Please take my advice and spend some time this summer thinking about how you'll put it into action.