Witness update

Policy changes

In preparation for the upcoming hard fork I have reduced my feed discount from 8% to 3%.

Separately, I have reduced by SBD APR to from 8% to 6%.

Background

The fork will reduce conversion time from 7 days to 3 1/2 days. This will reduce risk of conversions. The reduction in risk is appropriately 1/sqrt(2) or about 30%. However, we can also consider that two rounds of conversions can be performed in one week instead of only one. This doubles the expected return-on-investment for any incentivized conversions. In addition, conversions are currently charged an implicit fee attributable to hyperinflation as a consequence of the fact that STEEM are not delivered until 7 days later, by which time STEEM will have hyperinflated by about 3.5% (a portion of this is expected to be incorporated into the median price calculation, but not all of it). Post-fork there will be no further hyperinflation. As a result of these changes and other direct and indirect effects of the fork (primarily increased liquidity), I believe (subjective) that 3% post-fork is approximately neutral policy-wise relative to 8% pre-fork. Since the Steem economy will function quite differently post-fork, we will have to reassess to determine if this resulting level of feed discount is performing its intended role (supporting SBD price and incentivizing conversions for debt reduction) and make any necessary adjustments at that time.

The reduction in APR is for two reasons. First, the white paper reasonably instructs that interest must be stopped when SBD trades above 1.00 USD. This has been occurring, albeit intermittently, for about two weeks. While I don't agree with abruptly stopping interest (outside of possible extreme circumstances), I do agree with steady reductions in APR as long as SBD continues to trade above 1.00 USD. The reduction from 8% to 6% is the second of my steady reductions, and I am intending to do more such reductions as long as SBD continues to trade above 1.00 USD.

Second, aside from stopping when SBD is over 1.00 USD, the white paper gives little guidance about reducing APR or feed discount, only when to increase them. In comparing the two policy mechanisms for influencing SBD price, the feed discount has the secondary effect of adding incentive to perform SBD conversions, reducing debt (an incentive which carries an immediate cost to stakeholders). The APR does not (higher APR is either neutral or indirectly supportive of higher debt). Therefore feed discount should ideally be used on a high debt situation; higher interest should ideally be used in a low debt situation. (This is consistent with the white paper's instructions on when to increase these policy variables.)

After considering the matter along these lines, I believe the following rules are appropriate, and missing from the white paper, addressing how and when to reduce the policy variables:

1. When debt ratio is high, as SBD approaches and reaches parity from below, existing interest should be reduced or stopped before or along with reducing the feed discount.

2. When debt ratio is low, as SBD approaches and reaches parity from below, an existing feed discount should be reduced or stopped before or along with reducing the APR.

That is, with a large discount in SBD price, it may be reasonable to not take measures that would further widen the discount (such as immediately reducing an existing feed discount or APR), even if that means maintaining some policies supportive of the price that aren't a good fit for the debt level.

For example, in a low debt situation, a feed discount that may have previously been put in place in a high debt situation can be left in place at first to avoid making the SBD price discount worse. Likewise, in a high debt situation, with a large SBD price discount, an APR that may have been previously put in place in a low debt situation can be left in place at first to avoid making the SBD price discount worse. In both cases the policy mismatched with the debt situation should be reduced or removed (first or concurrently) when SBD approaches parity.

The latter (high debt situation along with low SBD price having been entered with a preexisting interest rate) closely describes the current environment. My policy has been to reduce both the APR and feed discount concurrently as we have approached and reached parity with USD. However, in order to ensure that still-high debt continues to be reduced, further adjustments may be needed to either or both policies.

Rewards from this post (in whatever form received -- STEEM, SBD, and/or SP) will be converted to SBD and burned, as were previous witness post rewards. Proof of the burn will be posted as a comment reply