(Repeats with no changes. John Kemp is a Reuters market analyst. The views expressed are his own)
* Chart: tmsnrt.rs/2maNQ81
By John Kemp
LONDON, March 6 (Reuters) - U.S. oil drilling activity has surged but so far the impact on production has been limited because of the long delay in completing wells and reporting output.
The number of rigs drilling for oil has almost doubled since hitting a cyclical low at the end of May and is up by more than 50 percent compared with a year ago, according to oilfield services company Baker Hughes.
But output of crude and condensates has risen less than five percent since May and is still below the level at the corresponding point last year, according to data from the U.S. Energy Information Administration.
The link between rig count and production has never been strong because of wide differences in productivity between rig sets and crews, and between shale plays and even individual wells.
But the increase in drilling should eventually show up in a significant rise in production, most likely with a delay of 9 months or more (tmsnrt.rs/2maNQ81).
Even under optimistic assumptions, it can take a very long time from a decision to increase drilling to show up as an increase in output in the official data.
In most cases, decisions about drilling programmes are based on the level and change in oil prices over the previous 2-3 months.
From the moment a decision is made to drill an additional well, there can be a delay of 1-2 months before rig arrives on location while contracts are placed and the rig is moved.
Rigging up, actually drilling the well and then removing all the equipment from the site can easily take another month.
The arrival of the fracturing crew and other well completion services usually results in a further delay of 2-3 months.
So the well could start producing 4-6 months after the initial decision is taken provided there are no unusual problems.
But production records are published for full calendar months and output for the first month is likely to be for a fraction of the full 30-31 days; full production will not be recorded until the second month.
There will be a further delay before output is reported to regulators and a final delay before regulators publish aggregated numbers.
Given all these delays, the full impact of a decision to increase production is unlikely to show up in the official production data for 9 months or more, and is based on prices that are up to a year old.
Output reported now is the result of decisions taken in June 2016 or even earlier. More recent decisions taken in the autumn of 2016 and early 2017 will not show up in output until later.
The big increase in drilling reported in the second half of 2016 and the first two months of 2017 will not show up in output until the second half of 2017.
The delays in the system are an important source of short-term oil price instability because decision-makers must make choices based on information that reflects conditions up to a year earlier.
At the end of the last boom, prices started to fall in June 2014, the rig count began to decline in October 2014, but output did not start to drop until May 2015, and those numbers were not reported until July 2015.
More recently, the doubling of oil prices since February 2016 has already triggered a big increase in U.S. oil drilling but its impact on production will only be known later this year.
In the meantime, OPEC must make a decision in May about whether to extend its production curbs without knowing how much U.S. shale production is already set to grow later in the year.
Editing by David Evans