We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Business Intelligence and the Elmendorf Rule

Originally published August 9, 2011

Douglas Elmendorf is the Director of the Congressional Budget Office (CBO). He has been in this post since early 2009 after serving for a few years as a senior fellow at the Brookings Institution.

Now, thanks to Charles Krauthammer, a nationally syndicated political commentator, he has been immortalized as the creator of the Elmendorf Rule. You may or may not agree with Krauthammer’s positions in general, but he tends to do a good job of laying out his arguments in favor of or against an issue. In a Washington Post column (July 8, 2011) titled “The Elmendorf Rule,” Krauthammer expounds on the issue of the national debt ceiling and the Obama administration’s plan to address it.

The Elmendorf Rule states: 1) You cannot estimate speeches, and 2) You cannot estimate leaks.

Let me provide some background. On April 13, 2011, President Obama delivered a major speech outlining a plan to address the national debt problem. It was surely a political statement, as any such major speeches are prone to be, intended to convey intention and align support for a specific approach that the administration believed would eventually lower the national debt to a reasonable level.

On June 22, two months after the president’s speech, the Congressional Budget Office (CBO) released its “Long-Term Budget Outlook,” a document that drew a rather bleak picture of the impact of our massive debt on the economy and its ability to bounce back from the recession. The CBO Outlook projected that in 2021 the debt would be 101% of GDP(Gross Domestic Product), meaning that it would be larger than the economy as a whole. By 2037, the CBO projects the debt to be twice as large as the complete U.S. economy. Anyone who has ever had a credit card or a mortgage knows, of course, that the larger the amount you owe the more interest you have to pay. Well, if the CBO Outlook is accurate, 20 years from today the federal government will be making interest payments amounting to almost 10% of the national GDP. The CBO’s message implies that tough times loom ahead for the American taxpayer if we don’t attack this problem now.

At a hearing of the House Budget Committee the day after releasing its Outlook, Doug Elmendorf was asked what the CBO’s opinion was of the proposals presented in the April presidential address. His response was variant number one of the Elmendorf Rule: “We don’t estimate speeches.”

In his column, Krauthammer comments on what he refers to as “highly placed leaks,” suggesting that the president is “prepared to offer Social Security and Medicaid cuts.” To these he responds toward the end of the column with variant number two of the Elmendorf Rule: “We don’t estimate leaks.”

Let’s examine the premise of the Elmendorf Rule, since it is most relevant to business intelligence (BI) practitioners, especially those of us who live inside the Beltway. Can we estimate speeches? Can we estimate leaks? Ultimately, it was the second part of Elmendorf’s response to the House Budget Committee that was most relevant to BI. He said, “We need much more specificity than was provided in that speech.”

And, of course, he was right. Yet there is a lot that we can do with the BI toolkit that will assist decision makers attempting to make sense of emergent public policy.

In effect, public policy is not made instantaneously; rather, it is the result of significant discussion where different positions are debated in the public policy arena over a period of time. From that debate, in which issues and positions are examined, argued, attacked, defended and refined, specific programs emerge that are carried forward as possible solutions to problems. It is usually at this point when a proponent must provide a budget, or equivalent accounting document, addressing the cost of such a program. And for any national level initiative, it must be defined and presented so Congress can vote for it and appropriate the necessary funds to make it happen.

In one of my previous columns ("Business Intelligence and Decision-Support: Some Thoughts for the Public Sector", February 2005) I tried to explain the process by using privatization of Social Security, an issue that was hot at the time, as an example of how public policy is made. And I stated that “ultimately there will have to be millions of decisions made for hundreds of thousands of situations involving the establishment of thousands of specific programs, policies and/or practices dealing with the topic.” In fact, we would need answers to a large number of questions before we could run the numbers to determine exactly what the specific financial or budgetary implications would be. But while there are dangers in oversimplification as we go into implementation and production, there is often advantage in simplification as we start the process of analysis and design.

Speeches can serve as initial frameworks from which metamodels can be built. Assumptions can establish the base for initial values, ranges and domains for key variables. Refinement of these assumptions can occasionally come from leaks. Why not? The result of early analytics using a speech can serve the purpose of reality checks and be useful input – potentially very useful input – to the discussion and the debate.

The Elmendorf Rule is a good one in general, but I am inclined to believe that any business analyst worth his or her salt starts to quantify programs, projects, proposals and initiatives as early as possible with whatever data they have at hand. Further research and many assumptions will get the ball rolling and the analysis will fuel the debate.

Even Doug Elmendorf may have had no choice but to do some analysis himself. With a Harvard Ph.D. in economics, he has had many positions where early quantification and economic modeling were paramount. At some of his previous jobs at Treasury, the White House Council of Economic Advisors or as Assistant Director of the Division of Research and Statistics at the Federal Reserve Board, we know he worked on budget policy, Social Security, Medicare, national health care reform, financial markets, macroeconomic analysis and forecasting, and other topics.

It will be fascinating to see what he can do with his skill set when he has to start his analysis by estimating speeches, leaks and other artifacts of the public policy arena.

  • Dr. Ramon BarquinDr. Ramon Barquin

    Dr. Barquin is the President of Barquin International, a consulting firm, since 1994. He specializes in developing information systems strategies, particularly data warehousing, customer relationship management, business intelligence and knowledge management, for public and private sector enterprises. He has consulted for the U.S. Military, many government agencies and international governments and corporations.

    He had a long career in IBM with over 20 years covering both technical assignments and corporate management, including overseas postings and responsibilities. Afterwards he served as president of the Washington Consulting Group, where he had direct oversight for major U.S. Federal Government contracts.

    Dr. Barquin was elected a National Academy of Public Administration (NAPA) Fellow in 2012. He serves on the Cybersecurity Subcommittee of the Department of Homeland Security’s Data Privacy and Integrity Advisory Committee; is a Board Member of the Center for Internet Security and a member of the Steering Committee for the American Council for Technology-Industry Advisory Council’s (ACT-IAC) Quadrennial Government Technology Review Committee. He was also the co-founder and first president of The Data Warehousing Institute, and president of the Computer Ethics Institute. His PhD is from MIT. 

    Dr. Barquin can be reached at rbarquin@barquin.com.

    Editor's note: More articles from Dr. Barquin are available in the BeyeNETWORK's Government Channel


Recent articles by Dr. Ramon Barquin



Want to post a comment? Login or become a member today!

Posted August 16, 2011 by Richard Ordowich richord@comcast.net

As humans we have this innate desire to understand everything, explain everything and now with computers, quantify everything. We feel the need to develop one rule that controls the universe and now it is suggested we develop a rule for analyzing speeches.


We have relinquished control of our economy to algorithms within the computers that run Wall Street. And when they fail to prevent calamities like the recent financial crisis and subsequent flash crashes in the market we try to find algorithms to monitor the algorithms! We’ve relegated humans to watching the algorithms and when they run amok we are unable to respond quickly enough to prevent disaster.


Now that these algorithms run the economy (I refer to this as The Age of the Algorithm) we focus our attention on algorithms that measure public sentiment, speeches and other human behaviors because we now have data from social networks like Twitter and Facebook.


I think a better way to predict the outcomes of legislation is to look at trends. For example, the establishment of private pension plans across the globe. In the US we have the 401K while in Canada RRSP’s. These have been in existence for decades. So what do the existence of private pensions funds foretell? They signify the trend to push the responsibility for pensions to the individual and away from the government. Social security will become welfare for those who do not have private pension plans. Social security will change. Either through quasi privatization such as increasing individual pension plan contributions limits or other politically acceptable ways. You don’t need to analyze speeches or leaks to draw that conclusion. Nor are there rules that would discover this trend.

This trend will be followed by a similar trend in medical care; pushing the responsibility for spending to the individual. Rather than government or industry deciding what healthcare plan or services you get, the individual will have to choose. This will also push the responsibility for what level of care the individual obtains. Do I get the MRI or not? Do I take a drug that will potentially prolong my life for 3 months? It’s pay as you go or as you die.

These trends have been going on for decades and you don’t need any business intelligence tools to assess their impact or consequences. As a matter of fact, the BI tools can’t even derive this “intelligence” even if you gave it all the data. Analyzing speeches and data from social media is entertaining and amusing but I wouldn’t invest or make material business decision or write legislation based on this.

Finally, the speeches and rhetoric relating to the debt was fodder to take the focus off the fundamental problem; jobs. The government can’t do anything about creating jobs so the debt was an opportune distraction. The lack of jobs is related to demographics, foreign trade and numerous other trends that have occurred. The government did not act on these trends before a catalyst caused a crisis.  Analyzing speeches and leaks won’t expose these trends. There were no speeches from politicians warning us about that fact that jobs would be disappearing. This is not a topic politicians want to convey to their constituents. It hits too close to home. Debt however is abstract and a topic that can be debated for years using budgets and projections that obfuscate the real issues. This makes good fodder for politicians. Speeches are infotainment for elections but don’t really affect policies. To determine what will really happen, it is necessary to look at trends. Trends are not rule based so algorithms and BI tools will not work. As the cliché says; “in business intelligence, the intelligence is between the chair and the keyboard”.


Is this comment inappropriate? Click here to flag this comment.