The classic example here is using concrete or asphalt for roads. Building a four-lane highway in concrete costs something like a million dollars per mile, all told. Building a four-lane highway with asphalt costs something like half as much. But the problem is that the asphalt highway will wear out much more quickly and will require you to go in every year and patch potholes, whereas the concrete highway will be much less susceptible to potholes and won't need to be torn up and repaved every 5-7 years. So by spending the extra money to build in concrete today, you're saving yourself extra expenses later on.

Of course, asphalt is cheaper and easier to put down, so there is always the temptation to go ahead and build the asphalt road now and take credit for it in front of voters while leaving the problem of maintaining those roads for the next person in office.

  if anything I would guess spending more would lead to
  more expensive maintenance as well

The idea is to buy a $100 thing that lasts 10 years, instead of a $20 thing that lasts one year, thus halving your long-term spending.

Of course, not all more expensive products last longer, so careful attention to value for money is a must.

>>You can trade off future maintenance by spending more initially, or vice versa

> Umm is that really true ?

Maybe not universally, but often, yes. E.g., you can build using more expensive materials that will last longer/are more resistant to environmental decay.

Lowering maintenance costs is one of the things you can spend money on when building infrastructure, but not the only thing, so it depends on how you spend the money.

> Not magic - innovation.

Innovation is great, but in most cases requires new upfront expenses that have already been covered for existing solutions. We already have a fairly good amount of infrastructure for producing infrastructure, so usually innovation is one of the ways you can spend more upfront to reduce long term maintenance.