The National Endowment for the Arts administers the U.S. Government's Arts and Artifacts Indemnity Program on behalf of the Federal Council on the Arts and the Humanities (FCAH). The Indemnity Program was created by Congress in 1975 for the purpose of minimizing the costs of insuring international exhibitions. In December 2007, Congress expanded eligibility under the Program to include coverage for works of art owned by U.S. entities while on exhibition in the United States. The Program has indemnified more than 1000 exhibitions, saving the organizers nearly $335 million in insurance premiums.

As someone who was "involved" for over a decade, I will defend it with passion.  It is a brilliant scheme and should not be knocked, although it will be by those who seek to profit by selling commercial premiums - especially unnecessary ones such as for War Risk. 

GIS probably saves our museums and galleries well over £15 million a year in commercial insurance, without which, as others have pointed out, so many exhibitions would simply not be possible because of the enormous costs of commercial insurance.  This is a substantial public benefit and government auditors have given it a clean bill of health, for the time being at least 

However, it depends absolutely  on every single person involved at every stage of the process behaving impeccably and taking the greatest possible care of every single object and never running any careless risks.  All need to be aware that these days there is no special "contingency" fund, so that any large claim would fall to be deducted from the overall allocations to the sector.  Also borrowers should be quite certain that the objects to be borrowed have a clear and clean provenance - but that is another story. 

GIS also depends on owners (especially trustees) not insisting on unreasonably high valuations.  If they do and the Scheme's administrators insist on a more realistic and defensible valuation, then the borrowers are in the awkward position of being asked to take out top up, commercial insurance which strictly they are not meant to do - as Government and Gov't funded bodies should not be paying any commercial premiums.

There's no disagreement that government indemnity schemes make possible exhibitions that otherwise wouldn't take place.  But the 'saving' of $335m in US/£15m p.a. UK isn't a real saving.  It's the market value of services that the government is providing with no charge.  I think it's wrong to consider it nearly free just because there haven't yet been huge claims.  It's like saying Russian Roulette is really safe because I played it three times without dying, or that slot machines are a great investment because my mate won £20 on a £1 game.  The idemnity is a real cost that should be calculated as the probability of loss multiplied by the value of loss, and the question is how it should be allocated.  Currently the state seems just to write a blank cheque.  A really big claim could have a catastrophic impact on overall funding to the arts if the government re-balances the books.

It encourages museums to mount the most spectacular shows of the most valuable works, because it makes no difference to them whether they borrow something worth ten thousand or ten million.  Curators crow about how they've cleverly managed to negotiate loans of the most precious and most delicate paintings, when these loans are actually the most dangerous and shameful.  How is it decided that the goverment will provide a subsidy of, say, £15 a head to people visiting a Van Gogh show, £20 for Leonardo or a few pounds for Barocci?  The answer seems to be no one.  Tax money is being arbitrarily allocated to subsidise the most risky enterprises, simply because the costs aren't properly allocated and accounted for.  Meanwhile basic security measures are being cut back, and museums can't afford enough guards to stop morons scribbling on their paintings.  There is a much stronger case to be made for providing more funding to look after existing collections.  

It's a subsidy that appears to be free, because no one seems to be picking up a tab at the end.  But the cost is real, it incentivises the wrong behaviour, and subsidy would better be directed at basic preventative security.  And to give full disclosure - yes, I loved seeing Late Raphael and I'm looking forward to Young Van Dyck.  Maybe that makes me a hypocrite, but it doesn't undermine the case against indemnities.  

 As an aside, surely the Prado show was under-insured at EUR954m?  It included:

• Nine studies by Raphael for the Transfiguration.  One sold recently for ~35m EUR.  Not all are that good, but it included the Ashmolean study, which is the best of them all, so let's say 250m for the group.

• St Cecilia altarpiece - can't think of anything like it on the open market for well over a century.  More important even than anything in the Mellon purchases from the Hermitage.  Surely 200m?

• Baldassare Castiglione, Great Holy Family and Self Portrait with Friend - say 500m for the three, putting each at less than twice Munch's The Scream.

The other 4m then has to cover three or four other portraits, other Raphael paintings including St Michael, St John Baptist in Wilderness, Small Holy Family (etc...), the works by Giulio Romano & Penni, and all the other drawings.  That obviously excludes exhibits from the Prado's own collection, which wouldn't be covered.