Monday, June 21, 2010

Furloughs Theoretically Save Money but Empirically Cost Job Performance

by Professor Tony Wheeler
Associate Professor of Human Resource Management

The current government budgetary picture in many states, if not most, and in many localities looks fairly awful. I won't focus on any particular state's budget woes because you can read any newspaper from around the country to get the details. Governments have three broad courses of action to close budget shortfalls: 1) Cut spending, 2) Increase revenues (e.g., taxes), or 3) A combination of choices 1 and 2. National, state, and local politics, obviously, affects the choices that governments will choose. As I discovered while doing research with a colleague, these political pressures even affect the research that nobody academics like myself attempt to conduct.

Especially when you want to understand the impact of furloughs on state employees.

For political reasons I cannot disclose in which state my colleague and I conducted our research, as my colleague, who collected the data, had to agree state anonymity. He also received a phone call from a lawyer in the DOJ of the state where we collected the data telling him to stop data collection and consult a university lawyer (a.k.a., cease and desist). So much for "academic freedom," human subjects approval from a sanctioned institutional review board, and the permission to collect data from 4 agencies within the state government. Luckily, we were allowed to proceed with our study after political pressure was applied from on high.

This little story should tell you about the use of furloughs to help close budget deficits: Politics, not empiricists, make the decisions, and politics and empiricism might be mutually exclusive. Why did this particular state allow us to continue with our research? Because in the midst of a political race, one side wanted to know the results of our study and the other didn't. What could we, nobody academics, possibly find that politicians seeking office would want to highlight in a campaign?

Simply, we found that as a furlough day approaches, employee emotional exhaustion, the key component of burnout, increases while job performance decreases. But it gets worse. The increased exhaustion peaks on the furlough day, remains high for a couple of days, and finally returns to pre-furlough levels a week or so later. All the while, employee job performance decreases until the exhaustion clears.

To further examine what happens during the nearly 3 week period bracketing the furlough (10 days before, 7 after), we also looked at how these exhaustion/performance effects might differ among employees. We measured employee predisposition to ruminate (e.g., constantly think about bad events) and employee positive recovery behaviors (e.g., doing things to get mentally get away from work). Unsurprisingly to us, employees who ruminated more over the impending furlough became more emotionally exhausted and took longer to recover from this exhaustion; consequently, these employees often had the biggest decrease in their job performance. Conversely, employees who reported taking steps to "get away" from work indeed had lower exhaustion, not as large of a decrease in job performance in the run up to the furlough, and faster recovery after the furlough. It's worth noting, however, that even these employees experienced increased exhaustion and decreased performance, too. It just wasn't as bad as those who ruminated.

What do we think happened here? Let's be honest about what a furlough is: a pay cut disguised as a day off. The more someone thought about this, the more upset they became and the less motivated they were to perform their jobs. For those employees who actively sought ways to "get away" from their jobs, the less impact the furlough had on them.

We viewed furloughs as a personal resource drain, and we recommended giving employees choices in this furlough (as some states do but the state we examined did not): Take the pay cut and continue working (after all, deadlines don't take days off and many furloughed employees work from home on their furlough days), choose the day of the furlough akin to a floating holiday, or schedule furloughs around holidays or weekends. Again, some states allow for these choices, but the furlough remains what it remains: a pay cut. And the work doesn't wait.

I excitedly presented these results to my LRS/MBA 573 course on Organizational Staffing, and the discussion quickly devolved into a discussion on the worth(lessness?) of government employees. Aren't government employees depressed anyway? I found this interesting and indicative of the pernicious effects of staffing practices like furloughs. On the one hand, many Americans view government employees as lazy at best and utterly stupid at worst. On the other hand, furloughs achieve exactly the results that confirm these stereotypes. We want better government employees but do things through policy to achieve the exact opposite; yet another example of the Folly of Rewarding A, While Hoping for B.

Is there any wonder politicians might be interested in our research?

(In case you're wondering, I did remind the class that I am a government employee who liked my job and wasn't depressed, but I don't think that stopped them from thinking that government employees were lazy and stupid!)

What we could not examine was how our results might link to financial or productivity outcomes at a more macro level. Unfortunately, governments have fixed budgets and metrics like sales or revenue in a for-profit sense are meaningless. We suspect that the decreases in employee performance affect the economic productivity of government organizations, which makes us question how much money furloughs actually save the government. Furloughs might make the short-term bottom line look better but ultimately do long-term productivity damage. This will have to remain a suspicion until we can find a government that tracks productivity and allows us to collect data. Good luck to us.

No comments:

Post a Comment