就业报告对企业为什么不那么有用了

美国劳工统计局（BLS）开始公布就业数据至今已有100年了。上周五上午，就像过去的每个月一样，这个设在华盛顿、没有花架子而且“一直在战斗”的独立、技术官僚型部门公布了一组反映美国就业形势的新数据，引起了经济学家、生意人、政客和电视台评论人士的关注。

然而，这份月度就业报告已经今非昔比。

10多年来，私营行业就业数据收集和分析风生水起，BLS得到的资金却一直原地踏步。在大数据受到热捧的时代，该局一直未能实现创新，实际上还砍掉了一些重要的新项目。情况本不该如此。虽然民间机构在补充和分析就业数据方面有一席之地，但就业数据有公益性质，它们应该出自一个独立的政府机构。

国会克扣已经不多的就业统计数据预算对美国经济毫无裨益。

2009年以来BLS的资金一直呈下滑或停滞态势。2016财年该机构的预算为2370万美元，低于奥巴马政府基于现实预期提出的全额预算要求。BLS的购买力低于2001年前后的水平。也就是说它一直无法对自己的工作进行升级或调整，也无法在数据收集或分析方面进行创新。该局不得不砍掉的项目包括大规模裁员数据和国际劳动力比较，就应对衰退和确定就业政策最佳实施方案而言，这些项目都很重要。

BLS现任及前任官员都曾表示，他们迫切希望扩大BLS的工作范围，以便衡量独立承包商的崛起、全球供应链对经济的影响以及金融危机以来工资增长率的变化。但可用资金不足。独立承包商这个例子尤为突出。政府和私人组织在研究中估算，所谓的零工经济就业占美国劳动力的比重介于5%到三分之一之间，如果BLS不能获得更多资源，公众对此仍将一无所知。

资金少其实并不意外。在国会，确保统计数据可靠很难成为一条吸引人的理由。但它应该如此。良好的信息是良好决策的关键。

10年来，经济统计数据变得越发重要，原因很多。大衰退开启了一个对劳动力市场感到焦虑和困惑的时代。零工经济的崛起以及制造业的发展等变化让决策层、教育机构、职业培训机构和企业主很难弄清楚就业形势。正如《华尔街日报》最近报道的那样，大学在指导学生方面尤为困难，原因是职位名称和技能要求正以前所未有的速度发生变化，在工厂车间和办公室都是这样。

私营公司正在竭尽全力来满足需求。各家公司的经济学家、数据科学家、软件开发人员和营销高管都在设计软件，其目的不光是帮助客户分析自身业绩，他们还要帮助公众和决策层了解就业形势。举例来说，目前多家公司都在考虑如何分析职位描述并让它标准化，以便帮助求职者、公众和教育机构。作为人力资源软件公司首席经济学家，我把重点放在怎样用专有数据来提供比大多数现有公开数据更精致的细节上，比如按大城市来划分新工作和求职者（包括求职成功和不成功的）所在地以及求职者发现就业机会的渠道。我的目标是让统计数据更丰富，能为求职者、用人单位、教育机构和决策层所用。

私营领域的这些努力，包括编制专有数据和开发分析工具，无疑能为BLS的工作提供补充。但我们这些私营公司并不奢望自己能填补这个空缺。

就收集客观数据并将其整合为可供决策层使用的材料而言，绝对没有哪家公司或哪个团体的能力可以和联邦政府相提并论。就业统计数据为关键的公众职能提供支持，从设定利率到确定总统的工作日程都是如此。因此，纳税人应该为编制这些数据提供支持。

改善统计数据的理由可能获得民主、共和两党的共同支持。去年，宾夕法尼亚州共和党众议员麦克•菲茨帕特里克和密西根州自由派众议员约翰•科尼尔斯共同主持了一次众议院简报会，重点就是需要改善就业数据。虽然为联邦政府部门提供资金的问题传统上更能引起左派的共鸣，但右派也已经意识到统计部门需要改进工作，最好的证据就是唐纳德•特朗普最近声称实际失业率高达42%。

国会可以改变这种局面。在接下来几个月，议员们将审议2017财年预算，因而又有机会为BLS提供充足资金，以便该机构启动必要的升级。

对经济领域从业者来说，每个月的第一个周五仍是一个神圣的日子。但它的意义取决于BLS能提供什么样的数据。美国人再也不能不把就业数据当回事了。（财富中文网）

约翰•赖特是领先人力资源软件公司iCIMS首席经济学家和前美联储经济学家。

译者：Charlie

审校：詹妮

It’s been 100 years since America’s Bureau of Labor Statistics started publishing its jobs numbers. This morning—and with every revolution of the moon around the earth—the independent, technocratic, no-frills Washington office carries the torch, captivating economists, businesspeople, politicians, and TV pundits with a new trove of data about the state of America’s employment situation.

But the monthly jobs report isn’t what it used to be.

For more than a decade, BLS funding has flat-lined while private-sector employment data collection and analysis have boomed. In the vaunted age of big data, the Bureau is failing to innovate and actually cutting important new programs. It shouldn’t be this way. While the private sector has a role to play in supplementing and analyzing employment data, jobs numbers are a public good: They’re supposed to be produced by an independent government agency.

Congress isn’t doing our economy any favors by scrimping on the already-small budget for employment statistics.

Funding for BLS has been falling or stagnant since 2009. Funding in the fiscal year 2016 budget is $23.7 million below what the Obama Administration requested based on realistic estimates of what it would take to fully fund the agency. The Bureau’s purchasing power is lower than it was around 2001. This means it hasn’t been able to update or refine its measures or seek to innovate in data collection or analysis. The Bureau has already had to eliminate programs, including Mass Layoffs Statistics and International Labor Comparisons—initiatives that were important for responding to recessions and identifying best practices in labor policy.

Current and former BLS officials have declared their eagerness to expand the Bureau’s work to measure the rise of independent contractors, the impact of global supply chains on the economy, and the dynamics of wage growth since the financial crisis—but there’s not enough funding available. The example of independent contractors is particularly glaring. Studies by government and private organizations estimate so-called “gig economy” jobs at anywhere from 5% to a third of the American workforce, and society will remain in the dark without more resources at the BLS.

The paltry funding shouldn’t be surprising. Ensuring solid statistics is hardly a sexy cause on Capitol Hill. But it should be. Good information is essential for good policymaking.

In the past decade, economic statistics have become more important for a range of reasons. The Great Recession ushered in an age of angst and confusion regarding the labor market. Changes like the rise of the “gig economy” and the evolution of the manufacturing sector have left policymakers, educational institutions, workforce training bodies, and business owners struggling to make sense of the jobs situation. As the Wall Street Journal recentlyreported, universities in particular are struggling to orient students as job titles and skill requirements shift at an unprecedented pace—whether on the factory floor or in the cubicle.

Private sector players are doing their best to meet the need. Economists, data scientists, software developers, and marketing executives at diverse firms are designing software not only to help clients analyze their performance, but also to help the public and policymakers make sense of the employment landscape. Right now, for instance, multiple firms are thinking about how to analyze and standardize job descriptions in a way that would help both applicants and public and educational agencies. As chief economist for an HR software company, I’m focusing on how to use proprietary data to provide finer-grain details than what’s available in most publicly available data—such as the location of new jobs vs. candidates (both successful and unsuccessful) by metropolitan area, as well as the channel by which the applicant found the job. The goal is richer statistics, useful for job candidates, employers, educators, and policy makers.

The cumulative effect of these private sector efforts—including the development of proprietary data and analysis tools—will no doubt supplement BLS activity. But we in the private sector are under no illusion that we can fill the gap.

There’s simply no firm or consortium that can match the federal government’s ability to collect objective data and aggregate it into useful material for policymakers. Employment statistics support essential public functions—from setting interest rates to shaping presidential agendas—so taxpayers should support their development.

The cause of better statistics could win a bipartisan following. Last year, Pennsylvania Republican Mike Fitzpatrick and Michigan liberal John Conyers teamed up to host a Congressional briefing focused on the need for better employment numbers. While funding for federal agencies traditionally has more resonance on the left, we need look no further than Donald Trump’s recent remarks that the real unemployment rate is as high as 42% to see the right’s recognition that statistical agencies need to improve their game.

Congress can rectify the situation. In the coming months, lawmakers will be considering the fiscal year 2017 budget and will have another opportunity to fully fund BLS so the agency can start making necessary upgrades.

The first Friday of the month is still a hallowed day in the economics profession. But the date is only as meaningful as the data that the Bureau can provide. Americans can’t keep taking labor statistics for granted.

Josh Wright is a chief economist for iCIMS, a leading HR software firm, and a former Federal Reserve economist.