It's hard to really know why a given government takes decisions to loosen covid restrictions (they do tend to justify themselves exhaustively when they tighten restrictions).
Part of it is that there is, in democratic countries, an ongoing tension between disease control and political cost, especially whenever a given country/jurisdiction has a sizable proportion of voters opposing restrictions.
A government can choose to:
- do nothing much and stick to doing nothing much (South Dakota)
- be aggressive in limiting covid from the start (New Zealand)
- initially do little and then apply restrictions when things get out of hand (most of the others)
Especially in the third case, when there is a government intervention, it is often presented as a temporary, exceptional, measure to limit political costs.
Conversely, whenever the situation improves sufficiently at the moment, governments can feel pressure to lift those temporary restrictions.
Masks are somewhat vulnerable to this pressure as they are a direct imposition on individuals and are not, from the point of any given individual, that efficient at protecting that individual. Mask mandates are efficient at lowering overall R0 however.
In political terms, this boom and bust approach is not all that different from the way governments used to alternate loose and tight monetary policy. Loosening up tends to be a vote winner until the economy overheats and there is a problem requiring tightening the belt. This was known to cause problems but still happened time and again.
Most Western countries have learned, over time, to uncouple politicians from driving monetary policy and instead rely on nominally independent central banks.
This is not to say that we should emulate the central bank approach, only to note that this type of stop and go policy making can be expected of government response to cyclical issues, at least until they and the voters settle on how to address these problems.
Governments right now seem to spend a lot of time looking at what others are doing with covid and assessing medical outcomes and political costs of other countries' policies.
For example, after Macron came up with the not-mandatory-but-very-inconvenient-to-skip vaccine policy in July, a number of countries quickly followed suit in limiting access to public areas for unvaccinated people. Now that Austria is making vaccines mandatory, Germany is apparently thinking about it.
Likewise, expect governments to watch decisions to either maintain or do away with low-cost restrictions like masks. British Columbia, where I live, had a mask mandate late June, then dropped it until Delta made it necessary again in late August. I am not sure where voters stand on having dropped mask requirements but they certainly did not appreciate the partial lockdown we endured later on. Expect that authorities will probably think things over thoroughly before dropping that mask mandate one more time.
Covid will probably require extensive juggling of these medical vs economic vs political costs at least until it becomes a more easily treatable disease (better, proven, therapies) or until enough people are vaccinated which by now looks to require almost measles levels of vaccination (95%). These calculations are also affected by changing levels of resistance/approval towards restrictions in each electorate.