Dear colleagues, though I'm a few weeks too late I want to wish you a Happy New Year! As far as I understand there are currently two proposed models for the RIPE NCC charging scheme for the period 2027–2031. Model A is category based where Model B is formula based. Here are my thoughts: 1. I do not support Model B for the simple reason that I think a formula based model is too complicated and therefore intransparent. Members should be able to immediately see (on paper, not being required to use a calculator) how many fees they are required to pay. 2. I support keeping the flat fee model (only) in case the fees are not increased. If the flat fee is increased all few years due to higher costs, I do not support it any more because it will be a financial burden for smaller LIRs. For instance, I think it would be a imposition for smaller LIRs to pay €3,000 per year one day. The hosting business is not getting easier—it's getting harder. 3. If a category based scheme is used, I think many categories are better than less categories. The current Model A has 21 categories (counting the €500 category in case the LIR has no resources at all) which is good. However, I think Model A is extremely unfair: 4. First, from what I can see, the largest single allocation is IPv4 /10 and there is not a single IPv4 /8 or even /9 allocation. So it is doubtful that even the largest LIRs will fall into the uppermost categories. That means there are categories which from the beginning on will never be relevant. 5. Also there is a big gap between the smallest and the largest LIRs. That means that the vast majority of fees continue to be paid by the smallest, small and medium-sized LIRs whereas the large and largest LIRs still contribute only a small amount (in relative terms). For instance, I own about an IPv4 /19 equivalent (= 32 x /24s). With this new model, my fees are increased from €1,800 to €3,080 which is quite much. If I would have an IPv4 /13 equivalent I would not be a sole trader but own a large company. I would not have 32 x /24s but 2,048 x /24s which is 64 times as much—but with this new model I would only pay €6,488. And if I would have an IPv4 /10, which is equivalent to 16,384 (sixteen thousand) IPv4 /24s I would only pay €8,503 though I would have 512 times as much space. This is not fair. And there is no simple middle course. If the stated goal is to make the scheme more fair then the larger and largest LIRs unfortunately need to pay way more because they are the only group which is not fee sensitive. It makes (almost) no difference for them whereas it does make a huge difference for us. Thank you for your time. --Kayo