Well, if the correlation exists because government spending is a lagging/sticky indicator, and what's actually happening is that globalisation is hollowing out

Well, if the correlation exists because government spending is a lagging/sticky indicator, and what's actually happening is that globalisation is hollowing out the manufacturing/wage earning class,there's no evidence that it will reverse, but the causation of this phenomenon is the opposite direction: government expenditure is increasing as a percentage of the economy because the other private part is collapsing, not because government expenditure is crowding private expenditure. In this instance it is an early indicator of distress, but not necessarily the cause.