French Caldwell

A member of the Gartner Blog Network

French Caldwell
VP and Gartner Fellow
15 years at Gartner
19 years IT industry

French Caldwell is a vice president and Gartner Fellow in Gartner Research, where he leads governance, risk and compliance research. Mr. Caldwell also writes and presents on knowledge management. His research includes analysis of the impact… Read Full Bio

Coverage Areas:

A Worst Probable Case Scenario for U.S. Government Shutdown Planning

by French Caldwell  |  October 4, 2013  |  3 Comments

http://enyart.house.gov/sites/enyart.house.gov/files/styles/congress_home_page_feature_rotator/public/home_page_feature/ShutdownImage.jpg?itok=immAGMDw

If you follow the shutdown news, you’ll no doubt hear some talking heads saying the shutdown will last a few more days, and some cleverer  pundits starting to link the timing for the shutdown to the debt limit deadline of 17 October.  Business, government and IT executives need a scenario on which to base longer term decision making related to the shutdown, so I’ve put on my inside the beltway strategist hat and come up with one.  I hope it helps.

First, the background.  Without going into political issues at stake, it’s clear that Congress is hopelessly deadlocked.  Neither political party is in control of Congress, and both parties are split further by factions.  There’s no clear path forward for a bipartisan coalition to pass a continuing resolution to fund government operations, and to raise the debt limit.  We are in a constitutional crisis.

With this deadlock, consider the following probabilities:

1 — There will be no agreement on a continuing resolution before 15 October, the next payday for government employees (0.8 probability).

2 — The debt limit will not be raised by 15 October, two days before the ceiling is reached (0.5 probability).

3 — If the debt limit is not raised, the Department of the Treasury will identify further extraordinary means to extend the debt limit for at most two more weeks through the end of October (0.4 probability).

4 — Come 1 November the U.S. government will not pay some of its bills, perhaps things like Medicare reimbursements and utilities of some federal facilities (0.2 probability).

As the above probabilities play out, the following scenario becomes more likely:

  • Government agencies will be told to identify the absolute minimum skeleton crew required to maintain public safety and security.  Courts will shutdown.
  • Plans for the use of National Guard and other military units to protect abandoned federal facilities will be established.
  • The government will announce a prioritization plan that will focus on paying many items not affected by the continuing resolution such as social security, veterans benefits, and interest payments on government debt.
  • It is highly unlikely, a virtually nil probability, that the U.S. Government will default on its debt.  Rather than allow even a temporary default, it is likely that as the ultimate debt limit is approached, Congress will pass incremental increases in the debt ceiling to assure world markets of a commitment not to default.

Federal agencies should assume a prolonged governmental shutdown which may require mothballing some facilities that current continuity of operations (COOP) plans assume will stay operational. State and local governments, and businesses that depend on federal data and federal IT systems should assume that data will not be available for up to two months.

This scenario is not the absolute worst case.  Rather it is a worst probable case. In 2010 and 2011, the Belgian government was deadlocked for over a year, but a caretaker government was able to pass a budget and continue running governmental services until a new government could be formed.  The U.S. Government constitutionally does not have that flexibility and it is mammothly consequential to the global economy, both of which mitigate against a prolonged constitutional deadlock, but also magnify the shock of one.

3 Comments »

Category: public policy Strategic Planning Uncategorized     Tags:

3 responses so far ↓

  • 1 Justin Blake   October 4, 2013 at 7:42 pm

    Where do you extract these probabilities from, and are you one of the pundits?

  • 2 French Caldwell   October 5, 2013 at 7:08 am

    Well, of course, by definition I’m now one of the pundits. The probabilities are a scenario analysis tool. Having no other information or experience, a probability is by definition 0.5 — either it will happen or it will not. When I have additional information that would sway that probability one way or the other, then it is either higher or lower than 0.5. And then there is the time dimension as well when you look at sequenced periodic events — So, for example, if there is a 0.5 probability that government employees will not be paid on 15 Oct, then it is half that again on 1 Nov. And as you get closer to an event, the probability either gets higher or lower.

  • 3 Edward Carney   October 7, 2013 at 2:03 pm

    A probability of 0.5 might be arguable in the absence of other information, but, clearly, we have other information. That information is the record of behavior of the principals during other shutdowns. Rational people could argue for tweaks to these values (some, for example, might consider p(default) = 0.2 to be a bit low, given the extreme behaviors observed so far), but, on balance, they seem reasonable.

    There are other questions, such as the contingent probabilities. For example is the probability of default higher, given that a C.R. is not passed (number 1 does not occur)? The rocky shoals ahead seem clear; whither the helmsmen?