Russia
Russia Will Become a Failed State: Think Tank
The Atlantic Council has polled 167 ‘experts’ for their views on what the future holds for Russia and it’s not good

Key findings
According to their poll:
- Nearly half (46 percent) of respondents expect Russia to either become a failed state or break up by 2033.
- 40 percent of respondents expect Russia to break up internally by 2033 because of revolution, civil war, political disintegration, or some other reason.
OK, so what’s the Atlantic Council?
The Atlantic Council is an American think tank in the field of international affairs, favouring Atlanticism. It was founded in 1961. It has sixteen regional centers and functional programs related to international security and global economic prosperity. Its headquarters are in Washington, D.C.
It was set up in honour of General Brent Scowcroft. He was a United States Air Force officer who was a two-time United States National Security Advisor, first under U.S. President Gerald Ford and then under George H. W. Bush. He served as Military Assistant to President Richard Nixon and as Deputy Assistant to the President for National Security Affairs in the Nixon and Ford administrations.
I guess that the Council could be a bit biased perhaps, but what about the panel of ‘experts’ it polled? These polls are great for generating headlines (mea culpa), but do they mean anything? Can we attach any weight to their conclusions? I’ll come back to that later with my marketing hat on.
Is it reliable?
That question is too vague, I know, so let’s look at their process.
Methodology
A total of 167 experts shared their insights on what geopolitics, climate change, technological disruption, the global economy, social and political movements, and other domains could look like a decade from now.
60% were US citizens with 30 nationalities represented (non-US are mainly European). Ages were evenly spread over 35, skewed heavily male.
26% private sector, 21% academia, with the rest split almost evenly across government, non profit and independents/freelancers (aren’t they private sector)?
Questions:
- were they self-selecting? (sample error)
- were Russians included? (bias)
- was a standardised questionnaire used, and if so, were the question prompts neutral?
There are many other sources of error, not least the way the wind is blowing when Putin gets of bed in his bunker. Anyway, it sounds like a ropy process, but these exercises always are. I’ve designed and used a few myself in the past when I worked in professional services marketing.
The most useful aspect of this process is to repeat it regularly and look at how the weight of opinion moves. This is typically done with purchasing managers on a monthly basis and can provide useful economic insights. Sharp divisions of opinion are also good for headlines.
But this looks like it’s just an annual exercise and that’s not much use — it needs to be quarterly at least. Here I think it’s just a press story exercise.

