Anthony Albanese is right about one thing when it comes to public holidays. A decision to grant an extra public holiday should the Matildas win the World Cup is one for the states and territories. The Fair Work Act specifies only eight national public holidays. Any others have to be "declared or prescribed by or under a law of a state or territory". The prime minister doesn't get a look in. Except that he says he is going to put forward the idea of a public holiday for a Matildas win at Wednesday's meeting of national cabinet, and he expects the premiers and chief ministers to "fold like tents". One already has. NSW Premier Chris Minns says should the Matildas win in Sydney on Wednesday and go on to win the final in Sydney on Sunday he'll not only arrange a statewide holiday but also a massive parade to celebrate "what would be an amazing life-changing and unbelievable event in the state's history". It wouldn't be the first public holiday for a sporting event. Melbourne has a public holiday for the Melbourne Cup, South Australia (improbably) for the Adelaide Cup, and all of Victoria for the eve of the AFL grand final. And it wouldn't happen on the Monday following the game. Minns says it takes seven days to gazette a public holiday. To critics concerned about the cost of an extra day off, Minns asks: "can you imagine the kind of energy, economic excitement? It would be an explosion of economic activity, particularly for the CBD." It is unconscionable to talk about the cost of a public holiday without also talking about the benefit - what the Productivity Commission describes as the "genuine social benefit associated with widespread community engagement in events, especially on days of cultural or spiritual significance". These benefits are deeper and richer than those of ordinary annual leave in which individuals or families are away from work, but not the entire city or country. The commission - no fan of unlimited days off - points to evidence that "more shared days of leisure enrich the relationships of people with their friends and acquaintances, which then improves the quality of leisure on other days". It's easy to imagine that happening should the Matildas win. A national holiday would bring the nation together, at a cost. And very new international research makes it pretty clear that cost would be small. One thing it wouldn't do, despite loose talk, is dent productivity. Productivity is usually defined as production per hour worked. If the number of hours worked is cut, production per hour worked is likely to stay the same, or even increase if people work a bit harder the next day to catch up. It's what prime minister Bob Hawke was getting at the morning Australia won the America's Cup in 1983. Although he memorably declared that "any boss who sacks anyone for not turning up today is a bum," he added: "you have to work a bit harder the next day to make up". Far more of us are able to work a bit harder to make up than when Hawke made the suggestion. Back then one in six of us worked in manufacturing, often on production lines that moved at a constant pace without the ability to catch up. These days it's just 6 per cent. More of us work at desks. A back-of-the-envelope estimate of the production that would be lost, quoted as if it is authoritative by Opposition Leader Peter Dutton, is $2 billion. But while some businesses will produce less and perhaps sell less if there's an extra public holiday, others will sell more (as Minns points out). And if they have to pay penalty rates to do it, that's not actually an economic cost. In the language of economists, it's a transfer from employers and their customers to employees. Working out the net effect of an extra public holiday on gross domestic product requires ingenuity, because it's hard to know what would have happened to GDP without it. Late last year two economists from Harvard University and the University of Chile, Rodrigo Wagner and Lucas Rosso, presented a solution. They took advantage of the fact that in many countries certain holidays aren't moved when they fall on weekends, meaning in some years those countries have fewer days off work from holidays than others. Examining data from more than 200 countries over the two decades leading up to COVID, they determined that the net dent to GDP from an extra public holiday was only 20 per cent of the GDP that would have been produced that day. READ MORE: As they put it, this means "an 80 per cent recovery with respect to the GDP that would have been lost if the effect were exactly proportional". An awful lot of us do a bit more work to catch up after a holiday, or are in jobs where that doesn't matter, or get more business because it's a holiday. As the authors expected, the effects vary by industry. In manufacturing, only about half of the expected losses are recovered. In agriculture, which continues regardless of holidays, all the expected losses are recovered. Applied to Australia's GDP, Wagner and Rosso's results suggest a hit to production as low as $1 billion from an extra holiday. And it is worth saying again that's not a loss. In return, we would get extra leisure, and get to share it with others. The pair used their data to examine other things. They found that self-reported happiness climbed in the years there were extra holidays, and deaths (mainly from job-related accidents) fell. Like most of the things we like, holidays have costs, but they are probably lower than we have thought, and (at least in the case of a Matildas celebration) offset by rather nice benefits.