:cityscape: Who are the governors of NEAR city?

If NEAR was a city, its inner workings - the transactions, businesses, and users will be governed by people who help keep the city safe.

• These people are validators, who ensure that the transactions and interactions in the NEAR are valid, so that no one can spend more money than they have or steal other people’s assets.
• Validators help to secure the network through a specific consensus mechanism (specifically, Proof-of-Stake) to decide the validity of a transaction.
• NEAR city is decentralised, meaning that multiple validators come together in order to keep it safe and no single entity can take control of the city.
• Validators validate and execute transactions on NEAR and oversee other validators.

Role of Validators on NEAR

> The mission of NEAR is to empower people to manage their assets, data, and the power of governance. Validators support this mission by ensuring a single entity cannot control or manipulate the way a network governs itself or manages data. -

:thinking:What could go wrong?

• Problems can arise if too much power concentrates amongst a few validators.
• Validators are “powered” by the share of NEAR staked with them alone or the public → more NEAR staked with the validator, more stake in the network.
• Specifically, suppose a group of validators collude to take over 51% of the staked supplied/hash rate. In that case, they represent the majority of the governors’ power to decide what is valid or not with a 51% attack.
• For proof-of-stake networks , this number varies and can be 33% (For the case of NEAR).

:telescope:Scope of this Analysis

In this analysis, we will explore the state of governance on NEAR. Specifically, we will look at:

• The degree of decentralisation across validators
• The change in decentralisation over time
• NEAR staked and NEAR stakers

:cut_of_meat: NEAR Staking

How much NEAR has been staked from September 2021?

:cut_of_meat: Stakers

How many new stakers were there on NEAR? Are they still staking?

NEAR Tables are not completely backfilled at the moment, making accurate staking status limited to trace.

- Date of Analysis: 02 August 2022
- NEAR Tournament 
- Contact: Pinehearst#1947 

Decentralisation on NEAR

NEAR Tables are not completely backfilled at the moment, making accurate staking volumes challenging to estimate. The present analysis presents the estimated degree of decentralisation based on available on-chain data.

Quantifying Decentralisation

There are a few metrics to look at when exploring the degree of decentralisation in a system. The article by Lee & Srinivasan [1] explains the use and limitations of the GINI coefficient and Minimum Nakamoto Coefficient in measuring decentralisation. [1] Srinivasan, B. and Lee, L., 2022. Quantifying Decentralization. [online] Medium. Available at: [Accessed 1 August 2022].

The degree of centralisation on NEAR spiked in early 2022 and subsequently declined.

The estimated Nakamoto coefficient declined from 15 to 11 in early 2022. Subsequently, this has increased to 14 at the time of the analysis. The GINI coefficient rose sharply but declined over time.

The staked share of the top 10 users has increased over time while the proportion dominated by the top validator has declined slightly.

While % of validators required to perform 51% attack has also declined, it is also worthwhile to note that the total number of validators on NEAR has also increased. More validators have to collude in this case.

Minimum Nakamoto Coefficient==: “Nakamoto coefficient as the minimum number of entities in a given subsystem required to get to 51%*of the total capacity”

• At present 7 validators alone take over 33% of the staked value. Thus, if with 33% cut-off is the number required, the MNC will be = 7 In short, an MNC of 14 means that 14 validators have to collude to make a 51% attack possible
• In general, the more validators that are required to execute a 51% attack, the greater the decentralisation *This may vary depending on the blockchain

GINI: The Gini index on the blockchain measures the distribution of staked NEAR across all the validators.

• A GINI of 0 equates to perfect equality, where the amount of staked NEAR Is equal across all validators.
• Conversely, a GINI of 1 equates to perfect inequality, where all the NEAR staked is concentrated in 1 validator.

% Dominance of Top Validator: What % of the total NEAR staked was contributed by the top validator?

Currently, the top user contributes around ~5.6% of the total NEAR staked.

Minimum % Validators to Gain > 51%==: What proportion of validators are required to gain a 51% attack?

While the MNC tells the number of validators required, it does not tell what proportion of the total agents are there out there.

Approximately 13% of validators will need to collude to control 51% of the total stake.

Approach: With every delegation transaction (deposit_and_stake), the shares of NEAR staked to a validator will be updated in the transactions log. Here, we extract the latest shares reflected across the top validators, filtering for the current seat price of 185k: . As I have not discovered ways to identify active validators on-chain, we will include validators that meet the seat price as of their last balance update.

The degree of decentralisation on NEAR is still rather low, given that over 40% of the current stake can be controlled by 10 or fewer entities. From the Nakamoto coefficient, 14 validators alone would be sufficient to collude a 51% stake to attack the network. The GINI coefficient of ~0.6 corroborates our observation the stake of NEAR concentrates amongst a few validators.

The amount of NEAR staked has been steadily increasing from September 2021.

1. A total of 142M near was staked (net), accounting for withdrawals
2. There was only one week of a net decline in NEAR staked (week of 13 June)
3. The volume NEAR staked and withdrawn are not correlated with the transaction counts. For example, we see taking counts spiking up to 22.4k in 4 Oct 2021, but only 1M+ NEAR were staked in total, which is comparatively lower compared to other weeks
4. The median amount of NEAR staked is generally increasing from early 2022. The median amount of NEAR unstaked was rather erratic over time.

NEAR drew a total of 90.4k new stakers from September 2021.

The number of new stakers has declined from > 1k weekly to ~ 500 new stakers.

Over 64.2% of these stakers are currently still staking their NEAR. But 28.4% of the unstakers had <1 NEAR staked.

Furthermore, the majority of the stakers were users who staked less than 1k NEAR (80k+). 54k of these users still had their NEAR staked and a substantial proportion only had < 1 NEAR staked.