It’s theoretically possible that the cost of most items climb faster than inflation, as long as highly influential items compensate it. For example, let’s say a typical household consumes three items:

Prices listed per monthly consumption (bread might be cheaper, but you’re buying a lot of it). Note how the total inflation was lower than two of the items.

With that out of the way, odds are they aren’t exactly lying, but misleading you guys: it’s easy to “tweak” the numbers down by arbitrarily not including items that got way more expensive, or changing the metrics. In my example above, they could replace “ham” and “butter” with “sandwich fillers”, bias the data towards the items with costs that didn’t climb as much, and then claim inflation wasn’t a big deal.

Note this is actually worse than simply lying - lying makes the result bullshit; misleading makes the result and the process bullshit.