The divisibility of coins does not change the deflationary aspect of the currency.  As many time as you divide them, you're still experiencing a deflationary event if someone uses a Sweft attack.  As the miner accumulates more coins, more money is lost. 

The deflationary aspect of bitcoin destroys the vibrancy of the mining ecosystem that is necessary to support a cryptocurrency.

Design a crypto with a minimum 2% inflation and it can survive outside threats.

Let me explain why deflation is bad.

How much transaction fees will support a 2% inflation rate?

If tx fee is .5%, the amount of currency needed to move to get the equivalent of 2% inflation would be 4x the amount of btc in circulation.

Assume tx fee is .1%, the amount of currency you need to exchange to get the equivalent  of 2% inflation is 20x the amount of btc in circulation.

So we assume that when block reward is negligent, that transaction fees will support the economy.

Let's assume the attacker has 50% of network hash.
Let's assume .1% tx fee and 20x velocity of btc.
The miner would in this scenario destroy 1% of BTC a year.

Thus the problem is that without inflation the more BTC that are spent, the more prone to a prolonged deflationary attack.

Thus with a cap on btc we are achieving the opposite of what we want which is spending, not hording.

Why do we not favor hording?

Because miners must profit to secure the blockchain from attack.

Otherwise Sweft's law will make a 51% attack and fork much easier.